Medium Format Forum
Register a free account now!
If you are registered, you get access to the members only section, can participate in the buy & sell second hand forum and last but not least you can reserve your preferred username before someone else takes it.
largest print size on CFV back
- Thread starter mr_orange
- Start date
jotloob
Well-Known Member
paul
The CFV sensor size is 4080x4080 pixel . (37x37mm)
When you devide 4080 by 300 (printer resolution) you receive a factor of 13,6 .
Now multiply 37mm by 13,6 and you will recieve 503,2 mm .
So this is the maximum image you can print at a print resolution of 300 and no further interpolation . 50,3x50,3 cm . If you use stair interpolation you can go for larger images .
The CFV sensor size is 4080x4080 pixel . (37x37mm)
When you devide 4080 by 300 (printer resolution) you receive a factor of 13,6 .
Now multiply 37mm by 13,6 and you will recieve 503,2 mm .
So this is the maximum image you can print at a print resolution of 300 and no further interpolation . 50,3x50,3 cm . If you use stair interpolation you can go for larger images .
It isn't exactly just straight math.
It's a bit like film. ISO 100 can take more enlargement than ISO 400 given the same size and viewing distance for the print.
Prints done up from 1X (12"X12") to 2X interpolation (24X24) should use a 360 dpi resolution setting. Larger display prints can be done at half resolution that due to viewing distance... ( A master print maker provided this information on the Leica Users forum).
Workflow using Adobe Camera RAW while processing DNG conversions done in Flexcolor allows interpolation directly from the RAW DNG file which provides the greatest possible data going in.
In Adobe Camera RAW set the resolution to 360, the bit depth to 16, and select the + size closest to the final print size you want.
If you want to further enlarge in PhotoShop, or work from an original file from a Tiff made via Flexcolor, select Res&le Image to: Bicubic Smoother in the PS enlargement dialog box... then set the exact size wanted.
At ISO 100 or 200, a 3X or roughly 36"X36" print should be easily possible using Adobe Camera RAW 16 bit interpolation.
Make a layer to then sharpen the file, so you can reverse it for smaller prints later.
I've done 40" square prints with little to no penalty. Given viewing distance and use of 160 dpi resolution, even bigger is probably possible.
It's a bit like film. ISO 100 can take more enlargement than ISO 400 given the same size and viewing distance for the print.
Prints done up from 1X (12"X12") to 2X interpolation (24X24) should use a 360 dpi resolution setting. Larger display prints can be done at half resolution that due to viewing distance... ( A master print maker provided this information on the Leica Users forum).
Workflow using Adobe Camera RAW while processing DNG conversions done in Flexcolor allows interpolation directly from the RAW DNG file which provides the greatest possible data going in.
In Adobe Camera RAW set the resolution to 360, the bit depth to 16, and select the + size closest to the final print size you want.
If you want to further enlarge in PhotoShop, or work from an original file from a Tiff made via Flexcolor, select Res&le Image to: Bicubic Smoother in the PS enlargement dialog box... then set the exact size wanted.
At ISO 100 or 200, a 3X or roughly 36"X36" print should be easily possible using Adobe Camera RAW 16 bit interpolation.
Make a layer to then sharpen the file, so you can reverse it for smaller prints later.
I've done 40" square prints with little to no penalty. Given viewing distance and use of 160 dpi resolution, even bigger is probably possible.
Wilco, I did a 5 foot wide print for a wedding client that was taken at the Detroit Tiger's Comerica Baseball park. It was shot available light with the H2D/39 @ ISO 100 using a H/C 35/3.5 lens @ f/5.6.
Taking the crop I did into consideration, the print was the equivlant of about 6.5 feet wide. You could view it at 1 foot distance and I even made an 8X10 of them from waist up. I enlarged the display file using the technique mentioned above, except I didn't know about using a lower resolution for display prints at the time, so the final file was huge.
Obviously, they are HUGH Tiger fans, and this was taken just before the Tigers won the ch&ionship. They put it up in their den.
Taking the crop I did into consideration, the print was the equivlant of about 6.5 feet wide. You could view it at 1 foot distance and I even made an 8X10 of them from waist up. I enlarged the display file using the technique mentioned above, except I didn't know about using a lower resolution for display prints at the time, so the final file was huge.
Obviously, they are HUGH Tiger fans, and this was taken just before the Tigers won the ch&ionship. They put it up in their den.
Thanks mark, yet again very useful information; I've often been told the best way to enlarge is through adobe camera raw, but was not informed of this technique.
Do you no why there is no option, for resembling image (such as bicubic smoother) when converting raw files in adobe raw Photoshop, wouldn't that be best done at the
Do you no why there is no option, for resembling image (such as bicubic smoother) when converting raw files in adobe raw Photoshop, wouldn't that be best done at the
Paul, the dpi numbers I quoted were from a long thread made on the Leica Users Forum concerning inkjet printing. There were posts made by a Master Printer located in NYC who does fine art printing for some famous artists and photographers... David something. Some of the Leica people there immediately knew who he was, and were delighted he even bothered to post a message given his "status". If I get a chance I'll see if I can locate it and post it here.
His work flow was specific to maximizing output from Epson printers, and there's some long technical mumbo-jumbo that went along with it ... which just glassed over my eyes ... but I take his word for it because of the stuff he was printing, and who it was for.
He also did a thread on scanning, scolding film users for scrimping on scanning gear after spending all their cash on Leica M glass with their heart-stopping prices. That was a good one because I got to back him up having "put up or shut up" for an Imacon 949 ... (speaking of coronary inducing prices : -)
I think 12 meter racing yachts and polo would be cheaper past-times then high-end digital... LOL
His work flow was specific to maximizing output from Epson printers, and there's some long technical mumbo-jumbo that went along with it ... which just glassed over my eyes ... but I take his word for it because of the stuff he was printing, and who it was for.
He also did a thread on scanning, scolding film users for scrimping on scanning gear after spending all their cash on Leica M glass with their heart-stopping prices. That was a good one because I got to back him up having "put up or shut up" for an Imacon 949 ... (speaking of coronary inducing prices : -)
I think 12 meter racing yachts and polo would be cheaper past-times then high-end digital... LOL
Hi,
I see mentioned a "printer resolution" of 300....and 360PPI (Pixels Per Inch) to the printer. Both are valid, and depends on what printer you are talking about.
Epson printers typically print with a native DOT resolution of 1440/2880. But, you only get one dot of one color at one location at that resolution. It takes an "array" of dots to make up a color, and this is accomplished by a "dithering" algorithm. Even though a printer has a particular dot resolution, that is different than what resolution you want to send to it to get actual colors/graytones.
Typically, the Epson printers like even divisors of their native resolution, and 180PPI is typically the lowest you want to go for "normal" print sizes, with 360 typically showing (much) better results. Anything above 360, with little exception, is not used.
The Epson printer driver (at least when I analyzed this a few years ago) takes the data you send it, and res&les it to 360, then runs another algorithm on that data to get the dots it will use to make the colors you are asking it to. This is why sending it an even divisor of it's actual dot resolution typically yields the best results. The more you res&le the data, the more degraded it will become, and res&ling it only once (and not a second time in the driver) typically gives you better results.
BTW, this assumes you are not using the Piezography quad-tone printing driver, but the "regular" Epson driver. For Piezography, you want to send it as much resolution as you can give it, their algorithm is very good, and will use it...up to 720, and even if it is not an even divisor.
WRT the 300, that sounds like a commercial printer or screen resolution, not a typical user ink-jet printer (at least not Epson). So, the real answer is, it depends on the printer you are sending the data to, as to what the maximum acceptable size you are going to get is. Of course, what is acceptable is entirely subjective.
Typically, I judge output size by saying I'm going to send the minimum of 180 to the printer, and go from there. If I can send 360, I do it. Sometimes I ups&le or downs&le in PS to get to 360 (or 180), print and see what the results are. You really do need to experiment for what gives you the best results, as it is not only equipment dependant, but image dependant as well.
If anyone wants a further explanation of the mechanisms involved, I'm more than happy to elaborate.
Regards,
Austin
I see mentioned a "printer resolution" of 300....and 360PPI (Pixels Per Inch) to the printer. Both are valid, and depends on what printer you are talking about.
Epson printers typically print with a native DOT resolution of 1440/2880. But, you only get one dot of one color at one location at that resolution. It takes an "array" of dots to make up a color, and this is accomplished by a "dithering" algorithm. Even though a printer has a particular dot resolution, that is different than what resolution you want to send to it to get actual colors/graytones.
Typically, the Epson printers like even divisors of their native resolution, and 180PPI is typically the lowest you want to go for "normal" print sizes, with 360 typically showing (much) better results. Anything above 360, with little exception, is not used.
The Epson printer driver (at least when I analyzed this a few years ago) takes the data you send it, and res&les it to 360, then runs another algorithm on that data to get the dots it will use to make the colors you are asking it to. This is why sending it an even divisor of it's actual dot resolution typically yields the best results. The more you res&le the data, the more degraded it will become, and res&ling it only once (and not a second time in the driver) typically gives you better results.
BTW, this assumes you are not using the Piezography quad-tone printing driver, but the "regular" Epson driver. For Piezography, you want to send it as much resolution as you can give it, their algorithm is very good, and will use it...up to 720, and even if it is not an even divisor.
WRT the 300, that sounds like a commercial printer or screen resolution, not a typical user ink-jet printer (at least not Epson). So, the real answer is, it depends on the printer you are sending the data to, as to what the maximum acceptable size you are going to get is. Of course, what is acceptable is entirely subjective.
Typically, I judge output size by saying I'm going to send the minimum of 180 to the printer, and go from there. If I can send 360, I do it. Sometimes I ups&le or downs&le in PS to get to 360 (or 180), print and see what the results are. You really do need to experiment for what gives you the best results, as it is not only equipment dependant, but image dependant as well.
If anyone wants a further explanation of the mechanisms involved, I'm more than happy to elaborate.
Regards,
Austin
A
Andrewdyer
Just to add a little bit further to Austins nice and technical post...
When we are getting something printed on a commercial offset screen printer, using what is known as a half-tone screen, the rule is 2 pixels for every halftone screen 'spot'. (BTW. A 'Spot' is made of lots of dots but we won't go there - it starts to get confusing)
Generally a high quality print is made using a 150 Lines Per Inch screen - so therefor all that is needed to make that resoution is a 300 Pixel Per Inch image (150LPI x 2 = 300).
Any pixel information above that is really a waste of information.
Of course if you print an image at a higher halftone screen resolution such as 175 Lines per inch, then that would require a 350 ppi image from Photoshop.
regards
Andrew
When we are getting something printed on a commercial offset screen printer, using what is known as a half-tone screen, the rule is 2 pixels for every halftone screen 'spot'. (BTW. A 'Spot' is made of lots of dots but we won't go there - it starts to get confusing)
Generally a high quality print is made using a 150 Lines Per Inch screen - so therefor all that is needed to make that resoution is a 300 Pixel Per Inch image (150LPI x 2 = 300).
Any pixel information above that is really a waste of information.
Of course if you print an image at a higher halftone screen resolution such as 175 Lines per inch, then that would require a 350 ppi image from Photoshop.
regards
Andrew