[Leica] Digilux 2 - wait for 8mp???

[tedgrant at shaw.ca (Ted Grant)]

Richard explained a ton of "stuff" ;-)
Subject: Re: [Leica] Digilux 2 - wait for 8mp???


> Well Ted, my 1800000 DPI prints ain't going compare to the prints you are 
> making (BTW, just ordered "Women In Medicine" along with "The Big Dig." 
> Should be here Tuesday...). So if auto-everything works for you, certainly 
> I ain't going to tell you to do things differently!!<<<<<<<

Hi Richard,
I'll say thank you very much for your explanation and taking the time to 
present it. Unfortunately I'm not the brightest candle on the block when it 
comes to numbers and things as you've explained. However the most important 
part is what you said in this respect and confirmed for me.

>>>So if auto-everything works for you, certainly I ain't going to tell you 
>>>to do things differently!!<<

Therefore good sir, despite your outstanding explanation, I intend to 
continue as you confirmed...:

<<<<So if auto-everything works for you, certainly I ain't going to tell you 
to do things differently!!<<

And a great big thank you for ordering our book, "Women in Medicine. A 
Celebration of their Work." And you certainly wont be disappointed with 
Michael Hintlian's book, "The Big Dig." An excellent piece of 
photojournalism at it's best.

Thank you.

ted





>
> But for the gearheads, it's all math. Some math are fuzzy, but it's still 
> math. Digital scans and digital files are represented as X pixels by Y 
> pixels. For example, you digilux 2 max resolution is 2560 pixels x 1920 
> pixels or just under 5 millions pixels. The sensor itself has more sensor 
> elements but some are "wasted." Think about this - this is all the data 
> there is for the picture. JPG compression adds another dimension to this, 
> and scaling algorithm is another, but when you come right down to it, you 
> have max 5 million pieces of information. To add more technobabble to 
> this, most digi sensors are Bayer sensor meaning that some of those 5 
> million pieces of data are already created but lets not go there. For my 
> eyes only, Foevon pictures sux eggs :-)
>
> Anyway, when you print, you send a bunch of data to the printer driver, 
> and after much work from various drivers, color management converters 
> etc., eventually some dots have to be laid on the paper. Supposedly, each 
> printer has its "native" resolution, related to but distinct from what the 
> 720/1440/2880 sez on the outside of the box. This is where that fuzzy math 
> comes in. I am sure other math heads understand it a lot better than I do, 
> but some people sez if you scale your output to this native resolution, 
> then the printer software does not need to do this conversion. So the 
> choices are either you do it yourself with Photoshop and a) trust 
> Photoshop to do a good job, and b) able to see the conversion, or let the 
> printer software does it for you automatically. Most people claim the 
> former is a better approach since Photoshop is a pretty darned good 
> program.
>
> So for the Epson printers, supposedly if you can scale your picture to the 
> print size you want, with 360 DPI, then it will look the best. This is 
> what I do normally with my scanned slides since I have lots of data to 
> play with. With smaller image files, I either scale it up to 360 DPI, or 
> in the case I mentioned where I only have about 170 DPI for the print size 
> I want (11x14), I just scaled it up to 200 DPI and let the printer 
> software handles it. I wasn't doing a scientific study, I was just trying 
> to see how good it can look and I am certainly pleasantly surprised.
>
> Of course my apology if you understand these issues already.
>
>
> At 01:35 PM 11/20/2004, Ted Grant wrote:
>>What I don't understand is this:
>>>>>(so I was scaling DOWN) and then compared to a print of similar subject 
>>>>>taken using the
>>>Minolta A2 at ~5MP resolution (it can go up to 8 normally) so it was 
>>>upscaled from ~170 DPI to 200 DPI at the 11x14 resolution<<<<
>>
>>I understand scanning film and using the Digilux 2 and doing all my slide 
>>and B&W scans at 4000 and making whatever slide or print size from that. 
>>The digilux is always used in highest size JPEG as it takes for ever in 
>>RAW to shoot, so I just gave up on it.
>>
>>My printer is "set to 300" whatever print size is required, again most 
>>times 12"X18" and the results are quite amazingly fine quality. Different 
>>from a wet tray print, however I've given up on people when they make 
>>constant comparisons between wet print to an Epson 2200 print, as I see 
>>that basically comparing apples and oranges, I think. Yep when one 
>>compares they look different, but that doesn't mean one is better than the 
>>other. Because if you look at one in one room and then go to another room 
>>and look at the other print you can't help but think they both look 
>>fabulous. ;-) Well OK in my small digital mind anyway.
>>
>>Generally from film print size is 9" X whatever to make the print and 
>>printed on the 2200 at 300. Sometimes I set it at 200 if I think I've 
>>enlarged a section a tad much. But rarely do I see any difference.
>>
>>I save the 4000 scan as my main file then I make whatever print sizes I 
>>want later. However, quite a bit of what we print for final print is 12X18 
>>on 13X19 Epson paper.
>>
>>I never cease to be amazed at the quality of the prints, particularly 
>>colour as it's amazing. I have a very simple system as I do most of the 
>>settings on auto everything. ;-) And it always looks so good I never mess 
>>with them. However, I must admit I'm slowly playing more with slidie bars 
>>and watching what happens on the screen. Then when it looks... "cool!" ;-) 
>>I save and print. ;-)
>>
>>How all this happens and why? Man I don't have any idea at all. :-) I just 
>>do whatever that makes it look good. :-) Save & print. :-)
>>
>
> // richard (This email is for mailing lists. To reach me directly, please 
> use richard at imagecraft.com)
> _______________________________________________
> Leica Users Group.
> See http://leica-users.org/mailman/listinfo/lug for more information
>