Canon 5d mark iii improved- olpf removed for higher resolution

3hats said:
I have had some more time to both shoot and inspect footage and my initial disappointment has been mollified. I would say that overall, it was a very nice improvement. I doubt that I would be tempted to add any post sharpening at this point.
There is IR contamination to some degree as my footage is biased a few points red, I knew that going in, though.
Was it worth the cost both to get it done and nullify the warranty - who knows, but I don't regret getting it done.
I know you want to see test results to "scientifically" establish this but I have no interest in being, both, told what to do and how it should have been done differently (sounds like my day job) You wonder why James and Andrew have not responded to your requests for more proof, check out tone of the requests that have been made.
However, if anyone wants to use my camera for a few days to do some "serious" testing against another 5d3 please drop me a PM and we can discuss how to get that done.
thanks
Bill
Click to expand...

Don't even blame you, man. Not one bit.

I'll take flowers and grass, wides etc, deepfocused and shallow stuff if and when you get time.
 
3hats said:
(...)

I know you want to see test results to "scientifically" establish this but I have no interest in being, both, told what to do and how it should have been done differently (sounds like my day job) You wonder why James and Andrew have not responded to your requests for more proof, check out the tone of the requests that have been made.
However, if anyone wants to use my camera for a few days to do some "serious" testing against another 5d3 please drop me a PM and we can discuss how to get that done.
thanks
Bill
Click to expand...
LOL Now I want to see the balcony or sofa disbelievers to move themselves to their scientific charts, after so generous offer ;) This community would be by all odds grateful for :-)
 
Congratulations, Bill. Very much looking forward to seeing some footage when you can post. Curious what the optimal settings will be with the camera configured like this.

This is NOT theoretical at all here. If this mod brings out the missing sharpness and if Canon released an all i-frame CBR version, I'm in. No doubt. The full frame thing for me isn't just about a depth of field thing, it's about using ALL the glass in a lens that was designed for full frame.

Also, we used to use a hot mirror on the red for an ir fix and it did help a lot.
 
@3hats
what maxmax use for ir/uv replacement is not 100% same as CANON stock ...there are a lot of people on photo forums who already done conversion and what i see their advice is to use custom WB....
 
Looks sharper than stock, thanks for the test Bill. A quick resolution chart test would be very helpful.
 
Even without the comparisons...

That's pretty convincing to my eyes, especially three minutes worth compressed to 200mb.

It's still soft, but not the same kind of soft as before the OLPF.

Thanks a lot, Bill.
 
3hats said:
I just posted some "test" footage on Vimeo
http://vimeo.com/40789276
thanks
Bill
Click to expand...


Looks good.

Definitely better than stock. Thanks Bill for being the next few who made the jump, it's very much appreciated. I'll add this to the front page too.
I downloaded the 1080P- Fantastic looking...

Pappas
http://PAPPASARTS.WORDPRESS.COM
http://TWITTER.COM/PAPPASARTS
BlackMagic Digital Camera Info Thread: http://www.dvxuser.com/V6/showthrea...mera-Technology-Announcements-amp-Latest-Info
 
Last edited:
Barry_Green said:
Why does "belief" even have to enter into it?

Look, it either IS or IS NOT sharper. This can be scientifically proven with a very simple test.

Why do people have to "believe" in it?
Click to expand...
Barry,

I used the word disbeliever in plural form ;) different than 'believer' or the verb 'to believe'. I also used figurative speech, yes. But the fact is that there has been posted some samples with and without the filter removed which some people didn't take them as proof of anything. Some people prefer 'to believe' in charts or as we wish, to make assumption in 'scientific' charts proof. To me and many others in this industry, charts can be misleading because can give you deceptive conclusions once many real world scenarios can't be reproduced by them.

E :-)
 
I used a Canon 24-105 lens at middle apertures for the test. Certainly not the sharpest lens on the block but one I use quite often. Shot on ALL-I processed with 5DRGB, simple CC in FCP 7.
if you have a suggestion on which resolution chart to shoot and the best method, (glass, aperture,distance) I will get that going.
bill
 
3hats said:
I used a Canon 24-105 lens at middle apertures for the test. Certainly not the sharpest lens on the block but one I use quite often. Shot on ALL-I processed with 5DRGB, simple CC in FCP 7.
if you have a suggestion on which resolution chart to shoot and the best method, (glass, aperture,distance) I will get that going.
bill
Click to expand...

Yeah I figured it was ALL-I, lots of compression artifacts. Try IPB it's a lot cleaner.
 
3hats said:
I used a Canon 24-105 lens at middle apertures for the test. Certainly not the sharpest lens on the block but one I use quite often. Shot on ALL-I processed with 5DRGB, simple CC in FCP 7.
if you have a suggestion on which resolution chart to shoot and the best method, (glass, aperture,distance) I will get that going.
bill
Click to expand...
Going with charts, at least, you can trade them with another 5DIII user (c'mon guys, open your inscription in this thread!) and this community can have finally both versions then, with and without OLPF :-)

Bill, you replaced both OLPF, didn't you?
 
Okay great. Looking forward to more and a chart (finally). Still looks like an improvement to my eyes from Andrew's video.

I think Emanuel is right, someone should be able to coordinate a test chart to compare with 3hats'. JCS perhaps?
 
3hats said:
I used a Canon 24-105 lens at middle apertures for the test. Certainly not the sharpest lens on the block but one I use quite often. Shot on ALL-I processed with 5DRGB, simple CC in FCP 7.
if you have a suggestion on which resolution chart to shoot and the best method, (glass, aperture,distance) I will get that going.
bill
Click to expand...

Thanks for your information, I was wondering how much did it cost you to do the conversion? maxmax told me it costs $575 to remove ONE OLPF and keep the dust removal system, you removed both OLPFs and kept dust removal system, I assume it would be even more than that?
 
Test Chart: ISO 12233:
http://www.graphics.cornell.edu/~westin/misc/res-chart.html (Vector format (PDF)).
Jason's test was pretty good: https://vimeo.com/39517721
See his vimeo notes on how he created the chart. If you don't have a printer that can print the chart at full resolution, perhaps a nearby Kinkos or similar can print it for you (I printed just the 4:3 region and could only do a comparison test against the still using a 600DPI laser).

After the camera & lens are set up and focused on the chart, a still is taken for later downscaling for comparison. This will show the camera was in focus, and also shows the max possible quality with the camera and lens combination.

Sharpening and contrast enhancement in post can show more detail (see Jason's video). For the base test, showing raw (preferably IPB) side by side images from a stock 5D3 and modded would be helpful. Then showing how far each can be enhanced with sharpening and contrast adjustment. Many folks are curious (as am I) how sharpened stock footage looks against non-sharpened no-OLPF (as well as sharpened no-OLPF).

For the 24-105, perhaps F8 @ 85mm would be good (researching the MTF for various f-stops and focal lengths for the lens in choice would provide best settings). If you have a sharper zoom (e.g. 70-200 F2.8L II) or a sharp prime, that would improve the results a bit. I can clearly see the 70-200 F2.8L II is sharper than the 24-105 F4L, however I have not compared them with a resolution chart.

http://www.luminous-landscape.com/reviews/lenses/28-105.shtml (sharper on the tele end; he compared at F8 to 24-70 where both lenses were very sharp).
http://www.luminous-landscape.com/tutorials/understanding-series/understanding-mtf.shtml
 
Emanuel said:
Barry,

I used the word disbeliever in plural form ;) different than 'believer' or the verb 'to believe'. I also used figurative speech, yes. But the fact is that there has been posted some samples with and without the filter removed which some people didn't take them as proof of anything. Some people prefer 'to believe' in charts or as we wish, to make assumption in 'scientific' charts proof. To me and many others in this industry, charts can be misleading because can give you deceptive conclusions once many real world scenarios can't be reproduced by them.

E :-)
Click to expand...
When I first got my 5D2 over 3 years ago I would have sworn that it had better resolution than my XHA1 or HV30 until I saw some proper comparisons done then did some of my own & realised that among other things sharpening & contrast can produce false detail that gives the illusion of increased resolution. I am not prepared to trust my eyes (or anyone else's) when it comes to deciding which camera resolves more detail just by looking at one or two compressed, edited, & graded videos. Only proper evaluation with charts or proper side by side comparison can be trusted. It's because I know not only my eyes fool my brain but everyone's do the same. It's just illogical to look at a clip & say it has more detail than another & categorically that this is because the OLPF has been removed. It's literally impossible to draw that conclusion.

Unlike most of the people involved in this discussion I actually own a 5D3 & would be absolutely ecstatic if some simple surgery on my camera would improve the image quality with no side effects even if it did void my warranty. The evidence that I have seen thus far is not just inconclusive it's non-existent.
 
You must log in or register to reply here.
Back
Top