|
Also why does the action clip off the bottom 10% of the blacks? That's a... strange choice.
|
#
?
Sep 8, 2016 16:57
|
|
|
|
| # ? Apr 25, 2024 23:13 |
|
|
MrBlandAverage posted:Also why does the action clip off the bottom 10% of the blacks? That's a... strange choice. That's just a personal preference, I find that it helps reduce cyan color casts significantly without sacrificing hardly any detail. It adds a second curves layer to do this, so you can easily hide/show it to compare the difference.
|
|
#
?
Sep 8, 2016 16:59
|
|
|
|
|
#
?
Sep 8, 2016 17:02
|
|
|
It doesn't bother me that you don't like how I do things.
|
|
#
?
Sep 8, 2016 17:11
|
|
|
Ok, so here are the results I'm getting. Here's the scan. I just converted it to JPG, but otherwise it's untouched from the scanner.  Here it is after running the action with nothing else apart from flattening it and saving.  And here's hows far I had to move the sliders to get a usable result. 
|
|
#
?
Sep 8, 2016 17:49
|
|
|
Not sure what's going on with the action on your side but this was the result I got: And then just some basic WB adjustments in LR:  Your scan does look super orange, more than any Ektar I've seen personally, so I dunno. Post a screenshot of your scan settings?
|
|
#
?
Sep 8, 2016 18:04
|
|
|
Huh, it's almost like every roll is different and you can't treat it with a catchall approach. Crazy
|
|
#
?
Sep 8, 2016 18:27
|
|
|
BANME.sh posted:
Scan looks fine mate. He needs to not use Lightroom for colour correction and use the PS curves tool to set the mid point. Also he needs not use your action - setting the level b and w points for channels is dead simple and more accurate and you learn more by doing it manually. e: set the b and w manually for each chan, used another levels layer to get the mid point and it all took 10 seconds.  Just add contrast and fine tuning like dukeku's edit and youre done. (i'm at work, hence PS elements) E: Setting the mid point gif 
bobmarleysghost fucked around with this message at 18:42 on Sep 8, 2016 |
|
#
?
Sep 8, 2016 18:34
|
|
|
The scan isn't fine - the red channel is clipped, which makes it not the best starting point. That said, dealing with less than perfect starting points is a good reason to know how to use curves. Here's a more neutral interpretation and the curves I used. 
|
|
#
?
Sep 8, 2016 18:51
|
|
|
 I don't know how to do this reliably without relying on my eyes and taste e: fiddled with the green channel a bit. VelociBacon posted:Are you setting each channel's (RGB) black individually? It might help. You can use the RGB all-in-one for setting highlight clipping. Adjusting levels like that will make your scans more 'reliable' without having to play with each image too much. Yes but in the end I always adjust the "middle point" on one or more channels (I don't know how it's called) until I get a result I like unpacked robinhood fucked around with this message at 13:49 on Sep 12, 2016 |
|
#
?
Sep 12, 2016 13:21
|
|
|
unpacked robinhood posted:
Are you setting each channel's (RGB) black individually? It might help. You can use the RGB all-in-one for setting highlight clipping. Adjusting levels like that will make your scans more 'reliable' without having to play with each image too much.
|
|
#
?
Sep 12, 2016 13:42
|
|
|
I'd like to get a size-accurate scan of a small object (a rangefinder cam as below). Think it would work if I slapped it on a scanner bed alongside a nickel and using the nickel's diameter to get a relative diameter for the hole in the RF cam?
|
|
#
?
Sep 21, 2016 20:05
|
|
|
Paul MaudDib posted:I'd like to get a size-accurate scan of a small object (a rangefinder cam as below). Think it would work if I slapped it on a scanner bed alongside a nickel and using the nickel's diameter to get a relative diameter for the hole in the RF cam? Probably, but why not just print up a sheet of 1mm graph paper and use that? Put the part on the glass and the sheet face-down over the part, and you can get all the measurements you need, baked right in. It looks close enough to two-dimensional that deformation and perspective changes shouldn't matter.
|
|
#
?
Sep 22, 2016 01:42
|
|
|
Anyone here scan film with a DSLR macro rig? Is there a go-to suggestion for this?
|
|
#
?
Dec 12, 2016 23:32
|
|
|
I did for a while. I suggest buying a good lightbox with a full color spectrum, a solid tripod head, and shoot tethered. I was having endless issues with color correction, because my lightbox produced very blue light. Shooting tethered is probably the best idea because you can see your focus on a huge screen and not a small viewfinder or lcd. Autofocus won't matter, so just buy whatever good manual focus macro lens you need. In fact, probably preferable because of the long focus throw I used my 55mm micro-nikkor (maxes out at 1:2, suuuuper long focus throw) but I would probably suggest 100mm+. Having a wide aperture doesn't matter much either, since you'll probably be shooting at f8.
|
|
#
?
Dec 13, 2016 08:59
|
|
|
Just got a Plustek 8200i. The process for getting two scanners set up in Silverfast is insane but once I got it sorted the results are a huge improvement from my V800 with 35mm scans.
|
|
#
?
Dec 29, 2016 22:47
|
|
|
Wild EEPROM posted:I did for a while. I've found that a Durst M301 or M601 enlarger can be converted to a cheap but useful copy stand. With those two models of Durst, the condenser head is attached with a 1/4" threaded mount, so you can remove it and attach a ball head directly to the stand. Works great as a solid, easily adjustable platform for your digital camera, and makes doing multiple formats in one sitting very easy. The soft box is a good idea. For a light source, I traded out my iPad for an LED array covered with either a piece of acrylic or translucent white glass. I do the 'scanning' in a darkroom. It's still possible that I'm terrible at manual color correction, but it could also be the color temperature of the light. I set a custom white balance on my 5D2 based on the LED array, but I'm not sure if that's enough (again, could just be my own ineptness). After doing a lot of digital reproduction with this DSLR, I've found that it works most naturally with 35mm film. The sensor size matches up with the 24x36mm frame, so I get 100% 1:1 reproduction with a 100mm macro. 645 is the next natural step up, then 6x9. Now that I have the stand, though, I think it could be worthwhile to experiment with stitching together multiple exposures for 6x7 or 6x6 frames. Finally, what's kept me going back to a flatbed scanner, at least for medium format, is the limitation of the 5D2's sensor. Inverting a .CR2 kind of brings out the worst of its characteristics, i.e. banding noise in the erstwhile shadows. So I often end up with obvious digital garbage in a daylight sky. And overexposing to compensate risks destroying detail in the shadows of the final image. All experiments with making an 'HDR' image by bracketing multiple exposures have ended up looking like crap, but once again that could also be due to the lack of requisite photoshop knowledge. Someday I plan on getting another digital camera to experiment with. Right now I'm thinking about either an Olympus EM5II (for the high-res shot mode and good macro selection) or a Sigma DP Quattro. I've had promising results, using a fixed-lens Foveon sensor camera in place of the 5D2, but it doesn't have the magnification necessary to reach the same level of detail as the Canon 100mm macro. SMERSH Mouth fucked around with this message at 04:33 on Dec 31, 2016 |
|
#
?
Dec 31, 2016 04:30
|
|
|
I am scanning a bunch of slides for the first time and I am doing it identically the same as C-41 or BW where you scan as a positive with all enhancements turned off. But I am finding that they are coming out quite dark, although not clipped in the shadows. The histogram is very far to the left. The slides look perfectly fine and well exposed held up to a light source, but it's like my scanner's back light is too dim. I have to crank the exposure in lightroom by 1-1.5 stops to get them looking normal. Is this common?
|
|
#
?
Jan 5, 2017 17:57
|
|
|
Are you adjusting the scanning exposure? Even after a generous scan exposure adjustment a lot of the slides I've been doing have been clipped in the shadows. Especially the red channel. Get it as close as you can without clipping the highlights is about as much as you can do. Someone else might have some better insight tho.
|
|
#
?
Jan 5, 2017 21:24
|
|
|
post one
|
|
#
?
Jan 5, 2017 21:36
|
|
|
here's one - http://i.imgur.com/Psh8r4P.jpg Like I said it looks great held up to a light source and there's detail in the shadows when I crank them, but I am just wondering if this is how the scan is supposed to look or what I should be doing to correct it. It was metered correctly so I am pretty confident it's not that. I am using Epson Scan
|
|
#
?
Jan 5, 2017 22:48
|
|
|
Looks fine to me. Here it is with the white point adjusted and a slight bump upwards on curves.
|
|
#
?
Jan 5, 2017 23:01
|
|
|
OK fair enough, thanks. When you say a bump up in curves you mean a point exactly at the middle (127, 127) and then just move it up?
|
|
#
?
Jan 5, 2017 23:08
|
|
|
|
| # ? Apr 25, 2024 23:13 |
|
|
Play it by eye on each image but for the most part the center should be fine
|
|
#
?
Jan 6, 2017 00:08
|
|