NZDavid Posted July 17, 2009 Share #1 Posted July 17, 2009 Advertisement (gone after registration) I've just had slides processed and scanned at my local pro lab from a winter trip last week through the top of NZ's South Island. It was pretty cold out there! I used EB100, which I am assured is extremely similar to E100G. Projected and on the light box, the colors look nice and bright -- though perhaps not as bright as E100G. The problem is the scanned images. All of them have duller colors, and what is more troubling, strange horizontal lines like an old CRT TV screen. Is this the scanning? I'll take them back on Monday. As always, any help appreciated. Link to post Share on other sites More sharing options...
Advertisement Posted July 17, 2009 Posted July 17, 2009 Hi NZDavid, Take a look here Bad scan or what? Help!. I'm sure you'll find what you were looking for!
NZDavid Posted July 17, 2009 Author Share #2 Posted July 17, 2009 Resizing makes the hoar frost even worse! This may be better., Link to post Share on other sites More sharing options...
NZDavid Posted July 17, 2009 Author Share #3 Posted July 17, 2009 Still blurry, sorry. Looks a lot better on screen apart from colors and lines. Link to post Share on other sites More sharing options...
andybarton Posted July 17, 2009 Share #4 Posted July 17, 2009 If they project OK, then it must be the scan. The one with the pronounced horizontal lines would back that up, I'd say. Nice spot though Link to post Share on other sites More sharing options...
NZDavid Posted July 17, 2009 Author Share #5 Posted July 17, 2009 Thanks Andy. I've just done some more research and it definitely looks like their scanner is at fault. Possibly a faulty power cube or carriage, whatever that means. FAQ 1 - Tech Support I'v encountered the same kind of lines when printing before, and that as the printer's fault. Scans are disappointing. Especially as scanning is expensive. Grr. Perhaps I should do my own...but results can be way off. One reason I still like slides: they are as close to the original as you can get. Link to post Share on other sites More sharing options...
acanalda Posted July 17, 2009 Share #6 Posted July 17, 2009 Probably this sounds a little stupid but....Do you have a mobile phone ? Have you scanned those slides with the mobile phone near the scan? I have experienced the same problem with my coolscan 8000 and YES...my GSM phone caused the lines. It may also occur with a home wireless phone (137 and 900 Mhz band) near the scan. Regards, Arturo Link to post Share on other sites More sharing options...
NZDavid Posted July 18, 2009 Author Share #7 Posted July 18, 2009 Advertisement (gone after registration) No, scans were done at a lab. But yes, cell phones (and computers) do affect other equipment. Hold a compass next to either and the needle swings wildly. My watch battery stopped -- I think because of proximity to my cell phone. Link to post Share on other sites More sharing options...
pop Posted July 18, 2009 Share #8 Posted July 18, 2009 Can you show us a crop of a striped area in its original size so that we can discern the individual pixels? Link to post Share on other sites More sharing options...
NZDavid Posted July 18, 2009 Author Share #9 Posted July 18, 2009 Philipp, here you go. Scan @ 18MB. Cheers, David Link to post Share on other sites More sharing options...
pop Posted July 18, 2009 Share #10 Posted July 18, 2009 David - that's not quite what I had in mind. In order to scrutinize the striping or banding from close up, we need an unspoiled part of the picture from very "close up". Sending all of an 18MB image will not work because the forum will compress it to a block of cinders. If the original scan has a size of 18MB, you will need to upload part of the picture corresponding to roughly 1% of the complete picture, i.e. 10% of its width and 10% of its height, preferably from a part which is afflicted by the stripes. That's an area of roughly the size I have outlined in the picture. [ATTACH]152745[/ATTACH] By the way, does the striping show with every viewing program or just with some? Afterthought: Has there been a different color resolution of those images at any time, such as 12 or 16 bits per channel (resulting in 36 or 48 bit colors) instead of 8 bits per channel or 24 bit colors? Link to post Share on other sites More sharing options...
NZDavid Posted July 18, 2009 Author Share #11 Posted July 18, 2009 That was a crop, not full frame, but here you go. Original size of scan was 18MB, 3,100 x 2,000 pixels. I resized using Graphic Converter. Are other viewers different? Yes, image looks a lot better in iPhoto and Apple Preview. Don't know the answer to your last question as I had the scans done at a lab. Many thanks. Link to post Share on other sites More sharing options...
pop Posted July 19, 2009 Share #12 Posted July 19, 2009 David - I had a close look at several of your samples. The GIMP is quite useful here as it allows a view of the pixels magnified to eight times their size without any interpolation. This is what I could detect I can not discern any striping in areas with high frequencies of details (e.g. the gravel in front) All areas with low frequencies are striped: the forest, the water. In darker areas the striping appears to be more visible. The striping involves in all cases four rows of pixels: there are two rows of similar brightness followed by two rows of brighter but again similar values. Most areas also exhibit artefacts in the other direction which usually have widths of eight (mostly) and four (less often) pixels. I.e. there are rectangular or square blocks of pixels. A good example is the sky. Given your data of 18MB for an image of roughly 2000 by 3000 pixels, I would guess that the original scans are stored in an uncompressed format (such as TIF) with 3 bytes per pixel. That would suggest a color depth of 8 bits per RGB channel. Given the regularity of the stripe widths and arrangements, I would think that the cause lies either with the scanner or - more probably - with software. Are the stripes visible at all in the uncompressed original TIF files (if such thy are)? If so, the fault lies with the lab. If not, I would suggest trying another software for the conversion and compression of the images. The similar phenomen of striped prints has entirely different reasons. In ink jet printers it is most often caused by misaligned or clogged nozzles. Link to post Share on other sites More sharing options...
NZDavid Posted July 20, 2009 Author Share #13 Posted July 20, 2009 Thanks again Philipp, the Iab is re-scanning. I'll post new versions a little later. Link to post Share on other sites More sharing options...
NZDavid Posted July 22, 2009 Author Share #14 Posted July 22, 2009 Well, they have re-scanned, and also did tree test 18 x 12" prints as a test, which look very nice. Examples of the scan below. Btw, the scanner is a Noritsu, used for bulk not custom scans. Link to post Share on other sites More sharing options...
pop Posted July 22, 2009 Share #15 Posted July 22, 2009 Much nicer. Glad that this is settled. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.