Pentax countdown..........

[dougpeterson]

Unfortunately, this becomes a matter of who to believe.

Trust no one.

Go shoot a high contrast scene with a 65+ and a D40, crop the 65+ down to the D40 size, do your best to make both look as good as possible and compare.

Or for that matter do the same with an H25 (7 year old digital back) and a 1Ds III. Those have the same native resolution and the 1Ds III has a 6 year technology lead.

If you're anywhere near Miami or Atlanta we'll be happy to provide the gear. If not, your local dealer is surely happy to let you take such a comparison.

Let us know your results.

Doug

Doug Peterson (e-mail Me)
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Cambo, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 *| *Cell: 740.707.2183
Newsletter: Read Latest or Sign Up
RSS Feed: Subscribe
Buy Capture One at 10% off
Personal Work

 

[monza]

Half the pixels, yes, but how big are the pixels? Presumably much larger. And of course supposedly it's true 16 bit vs 14. That appears to be open. I don't see the point of 11 point AF other than to fill specs to entice those coming from the DSLR market. It's simply not needed in a camera supposedly for 'serious landscape and outdoor photographers' which they state is the market for which the product was developed.

 

[douglasf13]

Trust no one.

Go shoot a high contrast scene with a 65+ and a D40, crop the 65+ down to the D40 size, do your best to make both look as good as possible and compare.

Or for that matter do the same with an H25 (7 year old digital back) and a 1Ds III. Those have the same native resolution and the 1Ds III has a 6 year technology lead.

If you're anywhere near Miami or Atlanta we'll be happy to provide the gear. If not, your local dealer is surely happy to let you take such a comparison.

Let us know your results.

Doug

Are you saying crop a 6mp image out of the center of the P65+, or resize the P65+ down to 6mp?

This seems to be a bit like P&S sensors vs. 35mm. P&S sensors have had better quantum efficiency than 35mm sensors for a while (in general,) but the 35mm sensors produce better images by leveraging sensor size and pixel pitch. I believe that is "theSuede's" point.

 

[Godfrey]

Irrespective of how the camera actually performs it does seem to establish that a number of features that MF users have been asking for can, in fact, be delivered. Without driving up costs prohibitively. Just looked in my gear cabinet and find that I still have my Pentax 645 kit and lenses. Maybe they're now worth more than the cost of a paperweight

I sold what remained of my P645 kit (body, two lenses, magazines, etc) just before the 2009 Holidays and netted around $750-800 for it. I bet that as soon as the P645D is shipping, the two lenses alone will be worth more than that.

The Pentax medium format gear, both 6x7 and 645, was always a stunning value for money.

 

[bensonga]

The Pentax medium format gear, both 6x7 and 645, was always a stunning value for money.

+1! Glad I kept all my Pentax 67 lenses etc....now I've just got to find one of those P67->P645 adapters. Darn they are hard to come by!

Gary

 

[Jorgen Udvang]

Half the pixels, yes, but how big are the pixels? Presumably much larger. And of course supposedly it's true 16 bit vs 14. That appears to be open. I don't see the point of 11 point AF other than to fill specs to entice those coming from the DSLR market. It's simply not needed in a camera supposedly for 'serious landscape and outdoor photographers' which they state is the market for which the product was developed.

Outdoor photographers don't use AF? For me, this could even be an interesting travel camera under the right circumstances, and there are plenty of situations I can think of where the nature of the AF system can be a limitations. It was a limitation when I used an E-1 (3 focus points) for that purpose.

Has Pentax said that this camera is mainly for nature and outdoor? My guess is that anyone buying it will use it for whatever as long as size, high ISO and/or speed aren't limitations.

I couldn't care less what the size of the pixels are. What counts is what the client sees on print. If the 22MP sensor is an older model with older technology, I doubt that it gives higher quality per pixel, even with pixels that are twice as large. Technology evolves. 14 vs. 16 bit colour depth? So far, the only "documentation" of the advantages of 16 bits I've seen has been endless technical discussions that have limited value in real life.

 

[carstenw]

The point being that the statement that the P65+ has worse per pixel quality than a D40 is just absolute garbage. It surely must be based on something (the guy seems very intelligent), but it surely is not based on processing raw files from each.

I think his statement is very specific, not commenting on the image quality at the end of the pipeline. His more detailed explanation was something about the noise inherent in the signal (signal quality), I believe, which he then went on to explain was made up for by sheer number of pixels and pixel size, and so on. I don't think anyone would dispute that the P65+ results at the end of the pipeline are among the best in the world, at this time.

He tests cameras professionally, and hangs out with a guy who swaps A900 sensors into D3x cameras for fun, and another who writes a well-respected raw program, so I wouldn't dismiss his comments.

 

[dougpeterson]

Are you saying crop a 6mp image out of the center of the P65+, or resize the P65+ down to 6mp?

Crop. Comparing the per pixel quality of each.

I don't dispute his expertise or his honesty. But I also know what I see in the final raw files.

Therefore it can only be that his comments about noise and dynamic range are focused on one aspect of the chain (that I listed above) and not the cumulative effect of the entire system. His paragraph that I quoted seems focused specifically on the sensor - A/D convertor. Perhaps if his comments are correct then the difference in outcomes (per pixel) is from the effect of the dark frame hardware/firmware/software. Or perhaps the noise is different in character in a way that is more easily removed while keeping detail, tonal smoothness, and color accuracy. Maybe it is mostly the software. I don't know the relative contributions of each. And frankly, I don't care. All I care about is that when I reach into the files to grab highlight or shadow detail I can do so to a greater extent and with more pleasing/natural results (per pixel) from a P65+ raw in C1 than with any other equipment/workflow that I've used.

Doug Peterson (e-mail Me)
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Cambo, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 *| *Cell: 740.707.2183
Newsletter: Read Latest or Sign Up
RSS Feed: Subscribe
Buy Capture One at 10% off
Personal Work

 

[dougpeterson]

He tests cameras professionally, and hangs out with a guy who swaps A900 sensors into D3x cameras for fun, and another who writes a well-respected raw program, so I wouldn't dismiss his comments.

My initial comment was a bit off in tone from what I intended. (so an official apology to him and to the forum). I should not have said the comment was "garbage" simply that it does not bare even a passing resemblance to my real world experience of what can be pulled from the final raw files of each system*.

*again speaking of high-end Canon dSLRs here. I've never shot the specific D40 he mentions.

 

[thomas]

... focused on one aspect of the chain (that I listed above) and not the cumulative effect of the entire system....

agree!
All these "RAW" comparisions are pretty pointless (IMHO). We simply don't know (and probably will never know) how much pre-processing is done at the RAW stage in-camera ... with Nikon, Canon, Sony etc. But we do know that DBs are too "primitive" to pre-process in camera... they need the manufacturers software to utilize the proprietary data. Hence the final result is what we have to look at.
If you look at the really phantastic clean files even an older DB delivers at base ISO you really wonder what all these guys are talking about. Simply doesn't add up...

 

[fotografz]

I think his statement is very specific, not commenting on the image quality at the end of the pipeline. His more detailed explanation was something about the noise inherent in the signal (signal quality), I believe, which he then went on to explain was made up for by sheer number of pixels and pixel size, and so on. I don't think anyone would dispute that the P65+ results at the end of the pipeline are among the best in the world, at this time.

He tests cameras professionally, and hangs out with a guy who swaps A900 sensors into D3x cameras for fun, and another who writes a well-respected raw program, so I wouldn't dismiss his comments.

I'm confused. Is there some actual application of something BEFORE it gets to the end of the pipeline?

I DO get what the gentleman is saying after wading through all the tech-no babble. :wtf: True science or not, interesting or not, once absorbed, it's just more data crammed into the old human hard-drive waiting its turn to be deleted as being 100% useless ... so, yeah, I'll eventually "dismiss it". Probably should learn to just not read it in the first place. My bad.

I agree with Doug, trust no one. My policy is to trust my two favorite evaluators ... my left eye and my right eye working together. Looking at Jack's P65+ pics ... I really don't care how the "sausage was made" ... I want my taste

 

[Guy Mancuso]

agree!
All these "RAW" comparisions are pretty pointless (IMHO). We simply don't know (and probably will never know) how much pre-processing is done at the RAW stage in-camera ... with Nikon, Canon, Sony etc. But we do know that DBs are too "primitive" to pre-process in camera... they need the manufacturers software to utilize the proprietary data. Hence the final result is what we have to look at.
If you look at the really phantastic clean files even an older DB delivers at base ISO you really wonder what all these guys are talking about. Simply doesn't add up...

This is the basic problem with Raws . DB backs are not being cooked before they hit the Raw processing engines and there maybe a lot of cooking going on before they hit on these DSLR files. You really don't know because i know C1 does so much to my Raw files. Simple case take one in ACR and it looks like crap bring the same file in C1 and it is a whole different ballgame. There IS a reason for software made specifically for the backs.

 

[PeterA]

This is the basic problem with Raws . ... There IS a reason for software made specifically for the backs.

+1 on that - so I wish I was smart enough to only use one system - but noooo not me nope...-

 

[douglasf13]

I agree with pretty much everything above. Like Mark said, no matter how "the sausage is made," the P65+ makes incredible images. My point in posting the thread from "theSuede" was simply to bring to light the possibility that the new Pentax's 14bits may not be much of an issue.

 

[tetsrfun]

Simple case take one in ACR and it looks like crap bring the same file in C1 and it is a whole different ballgame.
********
I never gave much credence to my RAW converter is better than X or Y until I did some direct comparisons between Phocus and ACR. It's not subtle. I suspect that it would be possible to get the colors "right" with some work in ACR but the difference in "detail vs noise" in underexposed shadows was huge. Of course the other possibility is "operator" error and/or ignorance.

Steve

 

[Guy Mancuso]

No the differences in noise floor, DR, color and everything else can be huge. My Phase files in ACR are 1.5 underexposed right at of the gun. Sure I can use it but I will not get the best out of those files from it.

 

[douglasf13]

No the differences in noise floor, DR, color and everything else can be huge. My Phase files in ACR are 1.5 underexposed right at of the gun. Sure I can use it but I will not get the best out of those files from it.

That's interesting, Guy. Usually ACR has a baseline exposure comp that artificially boosts the file, not the other way around. Have you looked at your RAWs in a program like RawAnalyze to see what your real RAW histogram looks like?

 

[Guy Mancuso]

No I have not and honestly been a C1 user for maybe 9 years and still have found nothing that can touch it and reason i bought a Phase to begin with but I am getting a chance to try the HD40 and Phocus soon here and see what that is like but again these are software programs built for these backs and they get the best from them. ACR ,LR Aperture are all nice programs but they are general cam programs . None of them are specialized for a certain camera.

 

[douglasf13]

Yeah, that makes sense to me. I used Leaf Capture when I had a Leaf DB. I'm just curious as to how much boosting C1 does to your Phase files compared to the actual RAW file itself.

 

[etrigan63]

No I have not and honestly been a C1 user for maybe 9 years and still have found nothing that can touch it and reason i bought a Phase to begin with but I am getting a chance to try the HD40 and Phocus soon here and see what that is like but again these are software programs built for these backs and they get the best from them. ACR ,LR Aperture are all nice programs but they are general cam programs . None of them are specialized for a certain camera.

Exactly, the RAW converters in question (C1, Phocus, etc) are essentially dedicated to the makers backs first (and in the case of C1) everyone else's second. Leica went with Lightroom since DNG is their native format and Adobe knows it better than anyone. I predict a lot of M8/M9/S2 tweaks in LR3 when it ships.