[Marc Sabatella]

Yes, of course. But just don't think 300 dpi is something that is of very high density so that our eyes cannot distinguish.

A 20" 4:3 LCD monitor in the resolution of 1600 x 1200 exactly has 100 pixels per inch but then one pixel consists of 3 dots, each of the R, G and B. So, physically it is already there 300 physical dots per inch for such a monitor

Um, OK, but when you computer 6.67x10" as the size at which a 6MP camera prints at 300dpi, that's 300 dots that already contain R, G, and B components. 900 dots if you wish to give each pixel credit for having three colors. That is, it's three times the resolution of your monitor. It really is a lot of resolution, and it really is true that most larger prints are done at less real resolution with no major loss in quality, because we don't view them as closely. Do all the math you want - even get it wrong, as you did above - but you can't change that fact.

The "line" can be drawn easily if we take the films we used as reference.

And why would you do that? Does something have to be as good as film to be of value?

[jay]

6MP is just too little by today's minimum standard, frankly - even most P&S DCs start from 10MP. In fact, the resolution difference can be clearly seen on a full screen (not full sized) picture displayed on a 19" monitor or larger. Furthermore, with such a low resolution, the user will have no room for further cropping.

Some of the best sports shots I've taken are on an old 1D I was given (well, "loaned"). It's got the flat Canon color -- but AF is obviously super snappy, fast burst rate -- a great camera for photojournalism.

It's ~4 MP, but I couldn't care less -- the chunk of glass that sits in front of the sensor is far more important.

A 6 MP sensor is, in my opinion, about perfect. You get gorgeous 300 DPI 8x10 prints -- if you ever need to go bigger, your DPI goes down, but unless you're hitting under 200 DPI, I doubt you'd be able to see a difference.

And, as mentioned, a 6 MP sensor has some big-ass photosites, which collect more light. What ever happened to ISO3200 when I sold my *ist DS and bought the brand new shiny K10D? Photosites shrunk, image quality went down.

Get a K100D Super (which supports SDM lenses). It's super small, very inexpensive, and, in my opinion, a GREAT camera to start out with -- and it will allow you to take full advantage of the latest, greatest lenses from Pentax.

[RiceHigh]

Um, OK, but when you computer 6.67x10" as the size at which a 6MP camera prints at 300dpi, that's 300 dots that already contain R, G, and B components. 900 dots if you wish to give each pixel credit for having three colors.

Not correct.

A 6MP camera only gives 300 colour dots per inch for *all* dots to be counted. What you got with a Bayler sensor with 300 pixels are actually just *physically* 300 pixels, not 900 as you supposed (there has never been 300 x 3 physical pixels to give your 300 pixels of colours). By considering the Bayler's artifacts that could come into play, a 6M picture from a 6MP DSLR printed at 10" x 6" cannot be much better than a 20" monitor at 1600 x 1200, both give you 100 dots per inch with complete colour information per dot, without any artifacts.

That is, it's three times the resolution of your monitor. It really is a lot of resolution, and it really is true that most larger prints are done at less real resolution with no major loss in quality, because we don't view them as closely.

If you have a DSLR which has 3 times the pixels along one side than a 6M DSLR, i.e., 9 times more pixels for the whole sensor area (a 54MP DSLR!), it is of course of enough resolution. However, your basic assumption was wrong from the beginning - One pixel of our DSLR is of *one* colour only. Getting the image with Bayler layout and re-use the pixels will not magically create 3 times more of *physical* pixels. As such, it could do no magic in getting more resolution.

Do all the math you want - even get it wrong, as you did above - but you can't change that fact.

I will be glad to stand corrected for what's being wrong on my part. The maths, as shown above, are simple.

And why would you do that? Does something have to be as good as film to be of value?

Film is not "not good" but it is just not as convenient and easy (to handle and for storage and for the workflow etc.). Film is the classic picture record media that we can easily refer to for the image quality. Fact is a typical consumer grade ISO 100 slide film has much resolving power than any of the current 135 or smaller format DSLR sensor, so there is still a long road to run in the digital technology until it can match film for just the resolution is concerned.

[jct us101]

Ken Rockwell? Is it you?

[RiceHigh]

I don't know about other's here, but I find what Rice High writes about interesting and informative.

Thanks, Mr. Coffee :-)

Perhaps I'm in the minority in my thoughts about Rice High's comments, however I do know I'm not in the minority in noting that this is a friendly place to talk about Pentax and photography in general - whether that be taking quick snaps of friends, professional studio work, or pixel peepers. Name calling and mudslinging are not needed and not welcome.

I don't think you are not the minority but there are fewer people who will stand out and speak.

I suggest that you put a check on your angst and perhaps view some of the other wonderful sections of the forum here. I'd like to suggest the Post Your Photo section to see some actual work by our Pentax brothers and sisters.
c[_]

I don't think any of them need to jump on me whenever I talk. No one forces them to read my posts and reply, especially when their replies are all off-topic and personal attack is just something that is not civilised and should not be welcome (and actually is not allowed according to the set forum rules). If they reply, I do hope that they do discuss more on the topic, or, just skip. Why is it so difficult? :-o

[Venturi]

How about K100D as my first Pentax DSLR?

DA 40mm should be a good choose, if I only want one prime lens right now?

Anyone else remembering what the original point of this thread is/was???

After 5 pages of debate on the merits of each others' opinions I wonder if bs6768 got his question answered, or if he got drenched in the megapixel piss, said phuque it and bought a Nikon.

If you're still reading, bs6768, my apologies for veering off track myself in one of the earlier posts and I do hope you've been able to get at least some useful information on Pentax DSLR options.

[lurchlarson]

Anyone else remembering what the original point of this thread is/was???

After 5 pages of debate on the merits of each others' opinions I wonder if bs6768 got his question answered, or if he got drenched in the megapixel piss...and bought a Nikon...

LOL!

[RiceHigh]

Anyone else remembering what the original point of this thread is/was???

After 5 pages of debate on the merits of each others' opinions I wonder if bs6768 got his question answered, or if he got drenched in the megapixel piss, said phuque it and bought a Nikon.

If you're still reading, bs6768, my apologies for veering off track myself in one of the earlier posts and I do hope you've been able to get at least some useful information on Pentax DSLR options.

So, thats what I suggested the K-m should be a better choice than a K100D right now for many reasons (I think RaduA has an excellent summary a few posts up on the differences), especially as resolution is concerned.

Of course, if he gets a K20D, he will get more resolution. But then the K20D may not be an entry level body with the lowest price in the *current* Pentax lineup then. The K-m is.

[jct us101]

Yes, before RiceHigh derailed the conversation, I hope he was able to get his own question answered. I still recommend the K100D if you want to pay $200 less than the KM. Of course you don't get the live view and smaller size, but it's still one hell of a camera.

[debbie]

.

If you have a DSLR which has 3 times the pixels along one side than a 6M DSLR, i.e., 9 times more pixels for the whole sensor area (a 54MP DSLR!), it is of course of enough resolution. However, your basic assumption was wrong from the beginning - One pixel of our DSLR is of *one* colour only. Getting the image with Bayler layout and re-use the pixels will not magically create 3 times more of *physical* pixels. As such, it could do no magic in getting more resolution.

.

so, are you saying that the K20D, for example, has a effective resolution of 14.6 Mp divided with 3??
I´m not with you here, maybe it is my english, I don´t know.

If you have a 300ppi print and still can see the dots, maybe the printer you are using
isn´t of the highest quality. Look at a 300ppi Lambda print and tell me you can see any dots. You can´t really compare prints with a monitor, the eye percives it very differently from a print.

[RiceHigh]

so, are you saying that the K20D, for example, has a effective resolution of 14.6 Mp divided with 3??
I´m not with you here, maybe it is my english, I don´t know.

Yes, in worst scenario, it is. In best scenario, it is still well below the idealistic 3104 LW/PH figure.

Realworld mesurement using the Imatest:-

ͬо²»Í¬ÖÊ ÈýÐÇGX-20Óë±öµÃK20D¶Ô±ÈÆÀ²â(ÏÂƪ)_ÍøÒ×ÊýÂë

They measure 1500 LW/PH as the best figure for RAW converted shots of the resolution chart, which is just less than half of the pixel figure in vertical run. In Jpeg, it could reach 2400 lw/ph at most but I am sure that the higher resolution is the trade off with more (colour) moire pattern - we can't get best of both world with Bayler sensors. Better demosaic algorithms with RAW convertor will decrease the resolution.

If you have a 300ppi print and still can see the dots, maybe the printer you are using isn´t of the highest quality. Look at a 300ppi Lambda print and tell me you can see any dots.

The problem here is not about the printer. Its just about the digital image does not have enough data to fit /feed into the printer for making all dots printed to be meaningful.

But when other people suggested that 300 dpi is just too high a printing density, I disagreed, I think it is just somewhat a standard setting.

You can´t really compare prints with a monitor, the eye percives it very differently from a print.

Our eye perceives them differently just because of the resolution difference, not because of the light is emitted from the source or reflected from it.

[Erik]

Yes, of course. But just don't think 300 dpi is something that is of very high density so that our eyes cannot distinguish.

A 20" 4:3 LCD monitor in the resolution of 1600 x 1200 exactly has 100 pixels per inch but then one pixel consists of 3 dots, each of the R, G and B. So, physically it is already there 300 physical dots per inch for such a monitor

You are extremely confused. It has absolutely no bearing on anything at all that a single LCD pixel has one R, one G and one B element, because when you're talking about real print resolution, what counts is one dot containing one colour. This is why your 1200 dpi printer does 250 dpi and no more -- it needs to rasterize every pixel into several drops of CMYK ink. In the same way, you can think of a monitor as being rasterized into drops of RGB.

In REAL printing, 300 dpi IS the highest quality anyone ever really uses, as I've said and as you have failed to disprove, because it is the truth. You're speaking complete nonsense. If you stopped talking so much about things you do not understand and actually made some REAL 300 dpi prints and LOOKED at them instead you wouldn't go around saying it is a low resolution.

[Erik]

So, we have to get at least 2.25 (square of 1.5) more pixels than the K20D before a DSLR can rival the resolution of just a roll of 135 ISO 100 (slide) film, which is 2.25 x 14.6 = 32.85 Megapixels!

With professional slides at lower ISOs, the lw/ph is 4800 and thus we need 4X more pixels to do the equivalence, i.e., 58.4 MPs!

You're wrong. The Sony A900 outdoes Velvia 100 in resolution, no question, and it is 24.6 MP. Although I don't have links to any proof, I suspect that it would outdo Velvia 50 similarly. Some very good print films (Ektar 100, Portra 160VC, TMax 100) are pretty bad in comparison -- although I haven't tried it, I would suspect that even the K20D sensor outdoes most 35mm print films available resolution-wise.

Look here: Technique - Ektar 100 test - Introduction and especially this page: http://www.boeringa.demon.nl/menu_technic_ektar100_resolution.htm

[RiceHigh]

You are extremely confused. It has absolutely no bearing on anything at all that a single LCD pixel has one R, one G and one B element, because when you're talking about real print resolution, what counts is one dot containing one colour. This is why your 1200 dpi printer does 250 dpi and no more -- it needs to rasterize every pixel into several drops of CMYK ink. In the same way, you can think of a monitor as being rasterized into drops of RGB.

I seldom use an inkjet printer to print photos. I use thermal printer which is designed to print photo only.

In REAL printing, 300 dpi IS the highest quality anyone ever really uses, as I've said and as you have failed to disprove, because it is the truth.

Disprove what? I consider 300 dpi is standard, not too high and not too low as a printing resolution. If you consider otherwise, that's your *opinion*, what do you want I to prove or "disprove"?

You're speaking complete nonsense. If you stopped talking so much about things you do not understand and actually made some REAL 300 dpi prints and LOOKED at them instead you wouldn't go around saying it is a low resolution.

I always print my photos at 300 dpi (as it is the default factory setting). I also found that my 13 MP images printed on a 4R thermal photo paper always look sharper than the 6 MP ones, with more details and resolution.

[RiceHigh]

This is not the first test which show such result, but I'm afraid that the bottle necks are the film scanners there, not the films.

If you are to compare *raw* resolution, one have to compare digital versus analog for using digital or analog workflow completely as a whole, e.g., a slide film to be processed analoguely and to project analoguely and then measure the resolution.

You can always do the reverse for digital and projected a digital image and print them on printed film and then do the measurement on resolution analoguely. By that time, it will not be surprised you will find that the film images are far superior!

You're wrong. The Sony A900 outdoes Velvia 100 in resolution, no question, and it is 24.6 MP. Although I don't have links to any proof, I suspect that it would outdo Velvia 50 similarly. Some very good print films (Ektar 100, Portra 160VC, TMax 100) are pretty bad in comparison -- although I haven't tried it, I would suspect that even the K20D sensor outdoes most 35mm print films available resolution-wise.

Look here: Technique - Ektar 100 test - Introduction and especially this page: Technique - Ektar 100 test - Resolution