Chasing Tack Sharp ... Just can't get there

[Pi]

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

I guess it mattered enough to you to buy an expensive MF camera.

The second shot has been worked slightly, nothing fancy just re-sample and sharpen

Resample to what resolution? What are we looking at?

[Pi]

I don't have a medium format digital camera! The 1Ds MkIII is not a medium format camera, as I said it is a six year old 135 format camera (FF), indeed it is essentially a pre release of the 5D MkII (not MkIII) sensor. The 1Ds MkIII can be bought for a touch over $2,000 now, the 5D MkII with the same sensor a fair bit less.

What you are looking at is a 7" crop from a 47" print. If you have a 23"-27" screen the crop should be close to 7" long, it depends on your screen resolution and size, if you have a tablet then zoom until the crop is 7" wide, you are then looking at a reasonably accurate low resolution representation of my 47" print.

My original file was 3744x5616, my resample which was as simple as Image Size, became 7488x11232, this is a four times enlargement, ie three out of four pixels are invented. This resample gives me a 31"x47" print at 240ppi.

All I did to get the crop I posted was from my print file I cropped a 7" section, I them downsampled that to 700px wide so it would display in the forum at the correct size. The paper copy needs more ppi than the screen and it gets it, On my monitor, a  27", the crop is 7" wide and it is a very accurate representation of a tiny part of the print. If you held my 47" print and my monitor next to each other the crop would be the same size.

So you upsampled it once, then you downsampled it by a factor that you cannot tell...  and you presented this as a proof that the 1DsIII can resolve as much as MF?

BTW, I confused you with ragmanjin, sorry for that.

[insanitybeard]

So you upsampled it once, then you downsampled it by a factor that you cannot tell...  and you presented this as a proof that the 1DsIII can resolve as much as MF?

BTW, I confused you with ragmanjin, sorry for that.

Where was privatebydesign stating that a 35mm size full frame sensor could resolve more/as much detail as medium format? As I read his text, his point was more about achieving pixel level sharpness, and that you don't need hugely expensive gear to realise it!

[Pi]

Where was privatebydesign stating that a 35mm size full frame sensor could resolve more/as much detail as medium format?

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

[Pi]

Sure I could have just shown the original file 1:1 [...]

This is what this thread is all about.

[Pi]

Where was privatebydesign stating that a 35mm size full frame sensor could resolve more/as much detail as medium format?

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

That is not making a claim for a 135 to be as good as a medium format, it is pointing out that you don't need medium format to get sharp.

I really don't understand how you could deduce your comment from mine, but there we gp, that is the internet for you.

If it is not a question of sensor size, then FF can as well as MF, right? Where is the logical mistake here?

[Cptn Rigo]

Hi RickSpringfield

Here are some advice to achieve your goal

1. Choose your sharpest lens, I believe that it is 50mm @ F4 (check it at photozone.de to find the best lens by resolution)
2. Find the lens "Sweet spot", generally is around F5.6 or F8
3. Focus manually with live view x 10
4. Use a remote shutter or 2-10 seconds shutter delay
5. Use mirror lockup to avoid vibrations
6. If you use external flashes, avoid flare
7. Use ISO 100 when possible

Done!

[insanitybeard]

Where was privatebydesign stating that a 35mm size full frame sensor could resolve more/as much detail as medium format?

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

And what part of your quoted text is stating that a full frame 35mm sensor can capture as much detail as medium format? 

[docholliday]

Another thing to look at too is the *quality* of the tripod and head. Shooting tiny things, like chocolates and jewels, I found once that the bogen head and aluminum tripod combo I had was actually amplifying slight vibrations in the room. Spent the good part of a week trying to figure out what was going on. AFMA was dead on, live view manual focus showed dead on, mirror locked up, wireless triggering with a PW, and was using the 100L IS - with IS off. Changed to an Arca-Swiss head, carbon Gitzo, then sandbagged the crap out of the tripod, lights, AND the product table. That took care of it.

It was because of the magnification and scale of the items. If the camera was a hair back further, the vibration disappeared, but I needed it to fill the frame because of reproduction size for maximum detail. Any closer to the subject, and the vibration disappeared, but then it was cropped in too far.

The product was lit with 5 strobe heads, large boxes and all. I suspect that if I didn't have the flash sync limitation at 1/250s (i.e., if I would have been using leaf shutters), I could have kicked it up a bit and been ok. It was just enough of a vibration that one could tell looking at the edges of the subject. And, it drove me nuts.

[Pi]

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

And what part of your quoted text is stating that a full frame 35mm sensor can capture as much detail as medium format? 

The second sentence, in the context of the discussion.

[insanitybeard]

Size doesn't matter anywhere near as much as most people think. This is not a question of sensor size, more a question of technique and light.

And what part of your quoted text is stating that a full frame 35mm sensor can capture as much detail as medium format? 

The second sentence, in the context of the discussion.

Pardon me, but I thought the topic was achieving sharp images. Is this the same as resolution?

[Pi]

Pardon me, but I thought the topic was achieving sharp images. Is this the same as resolution?

No, of course, you can sharpen to death and get a sharp image from any original.

Was that the topic of the discussion? 

[insanitybeard]

Pardon me, but I thought the topic was achieving sharp images. Is this the same as resolution?

No, of course, you can sharpen to death and get a sharp image from any original.

Was that the topic of the discussion? 

Nope, not to my knowledge. More about sharpness out of the camera without having to resort to sharpening in post.

[ksuweh]

This link might help you out. In the bottom left of the chart it states the average peak aperture of each lens/camera body combo.

http://www.reikan.co.uk/focalweb/index.php/online-tools/lenscamera-information/

[Jamesy]

This link might help you out. In the bottom left of the chart it states the average peak aperture of each lens/camera body combo.

http://www.reikan.co.uk/focalweb/index.php/online-tools/lenscamera-information/

Thanks for the link. Most of my lens are within range per the chart listed in this link. In some cases the Reikan aperture test showed my lenses as being better at wider apertures such as 5D3/70-200F4IS, mine is sharpest at 4.5 and the majority of people it is 6.3.

BTW, The Reikan Focal Pro is one of the best investments I have EVER made in my gear kit. It does what I would have a very hard time doing manually. It has changed the way I shoot - if aperture does not matter as much but sharpness does, I find myself selecting the Reikan defined sharpest aperture for the shot.

I also, incorrectly, thought that most lenses were sharpest around F8 - the Reikan sharpness tests showed me that is not the case - the sharpest areas in my limited testing seems to be in the widest 25% of F-stops.