I just started getting into scanning my own negatives and was looking for help on what I could do to get my scans sharper, If i can get them close to my lab scans i’ll be happy. Vlads test target helped with keeping the scans consistent not sure if i;m using it properly.
My current setup
Lab scan
I just want to chime in here before others jump in with workarounds… I think there really should be a default as such(lab scan) then tinker to your heart’s content because this query is probably the most requested in these forumns.
At the moment, none of the presets match or even come close to what a generic lab scan results in. I know that it’s all subjective and even the lab tinker with things(to their taste) but there is some sort of a “look” based on the qualities of the film that you get from a lab scan.
All of that said, the “look” of Portra 400, I believe(I’m no expert) is closer to that of your scan in this instance. Maybe with a little less saturation.
The title of the post is that your ‘scans’ aren’t sharp so since you have Vlad’s Test Target it might be better to start with that. Vlad’s Test Target is a very good guide to sharpness and resolution, with a good lens and a solid setup you should be discerning separation between the lines at Group -1 Element 6 in both the centre and corners at 26MP, just the centre with a slightly less good lens but the corners shouldn’t be far behind. With a very good lens you’ll get Group 0 Element 1 but the difference is unlikely to show up in an actual scan. If you are getting much less than that then there must be a problem somewhere.
Edit: I originally said ‘Group 1 Element 6’ - now corrected to ‘Group -1 Element 6’.
Oops we’re talking sharpness. I wrote in haste…
The primary cause of loss of sharpness when scanning with a digital camera is poor parallel alignment between the negative and the sensor.
How did you carry out this step? How do you use the Vlad’s chart? Normally, it should be easy to see if a portion of the chart lacks sharpness by comparing it to a similar portion. It’s normal to have a drop in sharpness between the center and the edges, but not between two opposite edges.
The second issue is the poor stability of the system. Do you use a timer or remote control to trigger the shot? Are there any unwanted vibrations during the capture?
I picked up the test target after reading about it here, I received mine less than a week ago so I’m not familiar with it yet. Off the top of my head i don’t remember the results i received when first testing it but i did test my lens from F2.8 to F22 I believe F8 gave me the best results. I will post my results here shortly.
As far as alignment, with my camera on the copy stand i use spirit levels on the camera, the holder, as well as using the mirror trick. Vlads chart? As in his 35mm test target strip or the alignment aid they sell? I can post my results I got with the test target.
I use the 2 second self timer on my camera to trigger each shot. My camera shakes a bit when capturing each shot but i believe it stables out before it reaches the timer but i could be wrong.
At least part of what you are seeing as the sharpness of your lab scans is the “unsharp mask” that the machines are automatically adding under the hood…
You should be able to get similar or better sharpness with a good sharpening workflow…
Hope that helps!
-Nate
I appreciate the tips this is what I came up with. It’s a step closer but comparing the flags from both scans the flag on mine is still soft.
I have downloaded your 4 images in order to compare them in Lightroom.I don’t know what size they were when you uploaded them but they are now all 1920 px on the long side and around 1280 px on the short. That isn’t enough to see how good the ultimate resolution according to Vlad’s Test Target is but to me it is enough to see that your setup is properly aligned as you say.
Your 26MP X-T30 will give you 6240 x 4160 px but your original scan really looks to have a problem to me, even at 1920 px. Compared to the lab scan at 1:1 in Lightroom it looks to me as if there is some vibration or movement creeping in and so no amount of sharpening is going to improve on it. In fact the ‘final edit’ image looks heavily over-sharpened with fringing all over the place.
The lab scan on the other hand looks very clean, no obvious problems from over-sharpening there, what was the original size in pixels of the lab scan?
I scan with an X-T2, I would expect to get a much better result than your original scan, so is it the lens, a focus issue, or is it vibration/movement, I think it’s the latter. As I’ve already said Vlad’s Test Target would show this up immediately. Perhaps it might be better if you could upload a cropped screenshot of the 1:1 comparison in Lightroom so that the image doesn’t resize.
I’ve uploaded my effort with the lab scan on the left, but naturally only taken from your 1920 px uploaded files so it doesn’t really work.
hmm weird my original scan the first one i uploaded is on my computer at 5655 x 3767 px, that’s correct i have another scan that’s exactly at 6240 x 4160. I only did the over sharpening on the final image to see if it helped based on some tips another user posted.
The lab scan came in at 3130 x 2075 at 72 dpi with a 24 bit depth. No clue if it matters. They used a Noritsu to scan.
You might be right with it being a vibration issue anytime i touch the camera it shakes violently, I thought with the 2 second timer it would be enough, I will have to try the 10 second timer, or try my tripod since it seems to be more stable than this copy stand. I noticed with Vlads test target the four circles on the inside aren’t focusing entirely based on the focus peaking i’m not sure if that matters? I could upload a picture if need be. I believe i understood the last part i’m new to scanning and the whole adobe ecosystem so i will upload as soon as i can.
After hours of fine tuning I got a sharper scan, not sharp enough but i think i’m content. I was messing with Lightrooms sharpening and seems i can get closer to my lab scan if i adjust it some more.
Thanks, it does matter really, it’s around 6.8 MP, don’t worry about the ‘72 dpi’, it’s not relevant, and 24 bit depth is standard, that’s 8-bits for each RGB channel. It’s a fairly standard size from a lab, and a useful size, you can use it for all sorts of things and larger scans from a lab would be a lot more expensive if they offered them. It is also a jpeg so all the adjustments that have been made to it are ‘baked’ in, including any sharpening as Nate has touched on.
BUT
With your camera you can get 26 MP images (6240 x 4160), much bigger and at the full bit depth that your RAW files allow. Don’t be put off by all these numbers but it should mean that you can get much better quality from your camera than these 6MP 8-bit jpegs from the lab. You will be working in Lightroom on your RAW file which is called non-destructive editing, nothing will change your RAW file and you can keep trying out things with virtual copies and comparing them.
So the concern is, for me, that there is still something wrong with your setup, you should not need to add a lot of sharpening in Lightroom to try and rescue your scan. Don’t be discouraged, such things are normal to begin with and once you’ve discovered what is causing this, and put in steps to prevent them from happening, then you should be able to get great scans from your X-T30.
Those rings on Vlad’s Test Target are called Siemens Stars and they are excellent, both for focus-peaking and for checking afterwards on your scan that focus is even across your frame. It is looking from what you say that it might not be and that is most likely alignment though with an unsuitable lens (and I don’t know your lens) you can get the situation where the centre is sharp but the outer part is out of focus.
Perhaps you could use a remote shutter release versus pressing the shutter button. That would eliminate the induced movement of the camera. If your camera is supported by Lightroom for tethered shooting, you could try that option.
I’m not sure if the xt30 can shoot tethered, I’ll look into it
Yea I agree I’m not sure what could be the issue at this point. i messed with my setup for a few hours trying to get siemens stars to align using focus peaking, the closest i got was the latest scan i uploaded. I’m using a Laowa 65mm f/2.8 2x Ultra Macro APO, I thought this was going to be the lens to use, Do you happen to have any recommendations on a good lens?
Do you have any recommendations going forward? I don’t wan to keep buying gear aimlessly trying to get super sharp scans and end up with a bunch of gear I don’t use. Thank you for your help so far this latest scan has me content but not sure if I’m content enough to stop asking for scans from my lab.
