[Welfl]

Focus peaking for stills can probably be added to any camera. But it might take a while before the peaking appears on the screen. Ie. the speed might not be satisfactory. That could be a reason why Pentax may not implement it on the K-5.

That's the best and most reasonable answer I've read on PentaxForums, so far. But..., read on. ...

Another thing of course is just business - Pentax won't make any more money by improving the K-5 now. They are probably just liquidating all their K-5 stock now, if they haven't already done so.

It's not common that businesses make functional improvements to products after their releases if there is no additional revenue stream

I've discussed this endlessly in many other threads here on PentaxForums and have gotten the same answers from other people, so I won't rehash it all here. So far, no one has convinced me to change my mind, but if someone eventually gives me a great answer I will be the first to admit I am wrong .

For example, here is a story from one year ago: Sony Firmware Update Brings Peaking To NEX-3 and NEX-5. Both of those cameras were introduced a few months before the K-5, which means they are even older than the K-5.

Oh well. Who knows? Maybe the various reviews of the K-30 will prove that other technological factors will help make its dynamic range and color depth nearly identical to that of the K-5, in spite of being only 12-bit RAW. It's not as if I am a heavy-duty photographer who absolutely needs those "two extra bits," although they sure would be nice to have, especially if I am going to spend this much money. Photography is just a pastime for me.

Eg. some subscription for new features in firmware upgrades for example. I don't know how many camera owners would pay for that.

I mentioned in another thread that I would be willing to pay for it, if it isn't more than a few dollars, because the Pentax software people have already developed focus peaking for the K-01 and the K-30 (and the K-3). It would only need to be tweaked a little to be compatible with the K-5's hardware. There would be no additional labor-and-time-intensive work that would need to be done for the sake of a soon-to-be-discontinued model.

If Pentax doesn't want to make that kind of investment, then perhaps they should open the platform a little to allow something like Magic lantern to be developed.

That would be great, but I don't have the luxury of risking buying a K-5 and then waiting for such an unlikely policy change to take place.

[Rondec]

12 bit versus 14 bit really isn't that big a deal. Most of the extra information isn't even usable and the file sizes are larger as a result. The only difference is probably a very minor one in dynamic range (using the K-01 versus K-5 performance on DXO Mark).

As to why Pentax hasn't added focus peaking. Well, there could be many reasons, but the biggest one is that, by all accounts, Pentax had planned to have the K5 sequel ready for sale now. It is close to two years old and traditionally, Pentax has had an eighteen month cycle. They don't have a large staff and those they have are working on newer and yet to be released products, not the K5.

I guess, I would favor Pentax offering a firmware that offered focus peaking (if it could be done) for a nominal charge (say 5 dollars). Then those who wanted it could choose to pay for it, those who don't care wouldn't have to. It could actually be a money maker for them too...

[loco]

AF is fast and accurate so far, at least as fast (if not a bit faster) as K5, under low light AF with K-30 is even more decisive. I can't comment on IQ yet, but the photos I took with the new camera are very good. Since it's still a preproduction unit, I can't post detailed comparison or comments on IQ or performance, yet.

As I said earlier, I don't mind getting either one between K-5 and K-30. The only thing I like better on K-5 over K-30 is the battery, and maybe the slightly better built quality.

Thanks Frank! Sounds good. I prefer using AA batteries so I really lean towards the K-30 for that. My hands are too small to even consider a K5 with grip. I appreciate all your info and pics!

[doozer]

That's the best and most reasonable answer I've read on PentaxForums, so far. But..., read on. ...



I've discussed this endlessly in many other threads here on PentaxForums and have gotten the same answers from other people, so I won't rehash it all here. So far, no one has convinced me to change my mind, but if someone eventually gives me a great answer I will be the first to admit I am wrong .

For example, here is a story from one year ago: Sony Firmware Update Brings Peaking To NEX-3 and NEX-5. Both of those cameras were introduced a few months before the K-5, which means they are even older than the K-5.

Oh well. Who knows? Maybe the various reviews of the K-30 will prove that other technological factors will help make its dynamic range and color depth nearly identical to that of the K-5, in spite of being only 12-bit RAW. It's not as if I am a heavy-duty photographer who absolutely needs those "two extra bits," although they sure would be nice to have, especially if I am going to spend this much money. Photography is just a pastime for me.



I mentioned in another thread that I would be willing to pay for it, if it isn't more than a few dollars, because the Pentax software people have already developed focus peaking for the K-01 and the K-30 (and the K-3). It would only need to be tweaked a little to be compatible with the K-5's hardware. There would be no additional labor-and-time-intensive work that would need to be done for the sake of a soon-to-be-discontinued model.



That would be great, but I don't have the luxury of risking buying a K-5 and then waiting for such an unlikely policy change to take place.

Regarding Focus Peaking: Besides the issue, that it might just not be possible to implement in the K5, I could think of quite a few reasons for Pentax not to do it. First, they have limited ressources. Even if you charge money for such an upgrade it, they simply might need those ressources for the development of the new products. To say that it cannot be that difficult to implement it in the K5 is a wild assumption on your behalf. Since the hardware is definitly slower, they might have to optimize the algorithms quite a lot to make it perform fast enough. Second, adding such a feature to a working firmware is always a risk from a software development perspective. You have to test it thoroughly again as it might affect other functionality. Third, I think you contradict yourself in that you say that focus peaking in the K5 would not hinder people from buying the K3 but than you are so eager to have this feature. Others were also very excited that the K-30 has it and there have been endless threads here and elsewhere that it should be added to the K5. This clearly shows that there is a high demand for this feature which then means that it is one piece of the puzzle to convince users to buy a new product. Not the only one, but not the least one as well...

So actually I see very little motivation for Pentax to include this feature in a product that is about to get replaced.

Regarding 12 bit vs 14 bit: Please show me a real life example for K-01 vs K-5 showing any difference in lowlight or dynamic range performance, that is noticable without pixel peeping (or at all actually). If you can't find one, then stop worrying about it.

[Welfl]

To say that it cannot be that difficult to implement it in the K5 is a wild assumption on your behalf.

It is probably a semi-educated and biased guess on my part, but I wouldn't call it a wild assumption. I've been a computer technician for a lot of years, so I think I am relatively familiar with the basics of hardware and software. That's not to say that I don't know what I'm talking about at certain times on certain subjects. There is always a very good chance that I don't -- and it may very well be true here --, but I usually try to base my assumptions on something more than just wild speculation.

Since the hardware is definitly slower, they might have to optimize the algorithms quite a lot to make it perform fast enough. Second, adding such a feature to a working firmware is always a risk from a software development perspective. You have to test it thoroughly again as it might affect other functionality.

This is one of the most reasonable and convincing theories I've seen so far. I am inclined to give you a strong benefit of the doubt, but it still remains for someone who works on firmware for a living to confirm it.

Third, I think you contradict yourself in that you say that focus peaking in the K5 would not hinder people from buying the K3 but than you are so eager to have this feature. Others were also very excited that the K-30 has it and there have been endless threads here and elsewhere that it should be added to the K5. This clearly shows that there is a high demand for this feature which then means that it is one piece of the puzzle to convince users to buy a new product. Not the only one, but not the least one as well...

Believe me. I have noticed this seeming contradiction myself and was expecting to have it pointed out to me sooner or later. It may or may not be a contradiction. I have repeatedly tried to make it clear that I do not own a DSLR yet; therefore, my "eagerness" doesn't count in my arguments. Furthermore, how many of the people who excitedly comment here (and on other forums) about focus peaking actually own a K-5, and how many of them actually own older and/or lower-level Pentax models and have already been planning to upgrade for other reasons, and focus peaking is just the icing on the cake? Speaking of which, mightn't some of these people be just as happy (or maybe even happier) to upgrade to a budget-priced K-5 with focus peaking instead of a similarly priced K-30? Either way, Pentax-Ricoh would profit from those upgrades.

In my previous comments about the pros and cons of adding focus peaking to the K-5 -- as far as Pentax-Ricoh's bottom line is concerned --, I have always tried to focus on present K-5 owners, not on myself. How many present K-5 owners worldwide even use old manual lenses? Are they just a niche in the total K-5 market? Or do they constitute a huge percentage? How many K-5 owners never use more than just the kit lens -- and never take it off of autofocus? How many never (or rarely) upgrade their firmware? How many refuse to use Live View, no matter what? There are definitely some people with that philosophy here on PentaxForums. How many use only AF prime lenses and would never dream of using manual lenses, or even AF kit lenses? Yes, I know. Many of the prime users sometimes (or maybe often) use manual focus with their prime lenses; however, many of them are also the very same people who refuse to use Live View. Of the percentage of K-5 users that do use manual lenses, how many do so purely for fun, and only on an intermittent basis? With all of that in mind, how many of the overall number of K-5 owners worldwide would be willing to "abandon" their expensive K-5s just to get focus peaking in a newer model? As much as I want focus peaking, I would never, under any circumstances, replace a very expensive camera that I had just purchased within the past two years in order to get it. I'm not that frivolous with my money, and I'm sure there are a lot of actual K-5 owners who feel the same way that I do, as much as Pentax-Ricoh might hope otherwise. For example, I have read a fair number of comments by K-7 owners who are very envious of the high ISO of the K-5; nonetheless, they have chosen not to upgrade to it because they say they spent such a large amount of money on their K-7s. I think we can all agree that a much better sensor is a far more compelling reason to upgrade than focus peaking is, yet these people chose not to upgrade anyway.

Regarding 12 bit vs 14 bit: Please show me a real life example for K-01 vs K-5 showing any difference in lowlight or dynamic range performance, that is noticable without pixel peeping (or at all actually). If you can't find one, then stop worrying about it.

The differences in these comparison shots may not be very dramatic to some people, but they are to me -- without pixel peeping (unless I am so caught up in this matter that I don't realize that I really am pixel peeping). I've scanned almost 10,000 negatives and photos in the past eight years. As such, dynamic range is a very important thing to me. I have tried (with a fair amount of success) to brighten the dark shadows and dim the bright highlights. I have studied dynamic range so closely over the years (as far as scanned images go) that the differences, as slight as they may be between the K-5 and the K-01, stand out to my eyes.

However, as I wrote in my previous comment (the one you quoted): "It's not as if I am a heavy-duty photographer who absolutely needs those 'two extra bits.' ... Photography is just a pastime for me." In other words, I probably should stop worrying about such a trivial issue since it will probably never affect the vast majority of the images I capture anyway. In almost every other respect, I am absolutely thrilled with the K-30.

[madbrain]

12 bit versus 14 bit really isn't that big a deal. Most of the extra information isn't even usable and the file sizes are larger as a result. The only difference is probably a very minor one in dynamic range (using the K-01 versus K-5 performance on DXO Mark).

That's likely indeed that it won't make much difference. Larger files also means slightly lower fps possibly.

[madbrain]

The differences in these comparison shots may not be very dramatic to some people, but they are to me -- without pixel peeping (unless I am so caught up in this matter that I don't realize that I really am pixel peeping).

That is not a comparison of 12 bits vs 14 bits, however. That is a comparison of two different cameras.
Also, the files in the first shot are all JPEG with 8 bits of data only. I don't see how you can conclude anything about the usefulness of the 2 extra bits.
IMO, the main way to do that would be to have a camera that has a way to shoot 12 and 14 bits and compare, or shoot at 14 bits then truncate the lower 2 bits. Then again, most displays only have 8 bits per color channel, so I'm not sure how you tell the difference visually anyway. I think it matters mostly in post-processing.

[Pentaxor]

question is, how bad does one really need the 14bit over 12 bit? if one uses such data info to validate one's work, then it's all worth it. but if one prefers 14bit for the sake of just having 14bit, doesn't make much sense. it's like owning a ferrari sports car that you only drive for 20mph.

[Pentaxor]

what I'm saying that the difference between 12bit and 14bit are negligible or non-existent for normal situation and use. too subtle of a difference to even get noticed.

[Welfl]

That is not a comparison of 12 bits vs 14 bits, however. That is a comparison of two different cameras.

I'm glad to hear that. It is encouraging. Let's hope then, that the K-30 is much closer to the K-5 than the K-01.

Also, the files in the first shot are all JPEG with 8 bits of data only. I don't see how you can conclude anything about the usefulness of the 2 extra bits.

Somewhere in Frank's long thread, he provided links to the actual RAW files (which you could download). Those links are probably dead by now. I forgot about that fact until you pointed out that those are JPEGs to which I linked. Anyway, I noticed the difference in quality in the actual RAW files too.

Nonetheless, I sincerely hope you are right about the comparison being between two cameras and not between 12 bits and 14 bits.

[Welfl]

question is, how bad does one really need the 14bit over 12 bit?

I already confessed that I probably don't need it as much as I think I do. That's why I concluded my mile-long comment above by saying pretty much that. In other words, I was attempting to come down to earth. I've actually seen some gorgeous real-world images that were produced by the K-01.

what I'm saying that the difference between 12bit and 14bit are negligible or non-existent for normal situation and use. too subtle of a difference to even get noticed.

There are people here on PentaxForums and elsehwere (including actual reviewers) who say (or at least imply) that those extra two bits make a meaningful difference; however, since I have never shot in RAW, I can only take their word for it and your word for it. For my sake, I sincerely hope you are the one who is right.

However, I must ask: If the differences are really as negligible as you say they are, why does Pentax treat 14-bit RAW as if it is only for higher-end pros? Why don't they also give it to the slightly more consumer-oriented K-30 and the K-01? Is the 14-bit-vs-12-bit debate being used as a cheap, deceptive marking gimmick? I wouldn't be surprised if it is. And it also cannot be argued that Pentax is taking consumers' and enthusiasts's hard drives into consideration, because most of them these days have just as much hard-drive space as the pros do, so file size cannot be used as an excuse. I, as a Mac technician, certainly have tons of hard-drive space at my disposal.

[creampuff]

Welfi, like Pentaxor, I think you're overthinking and fussing about the differences between 12 and 14 bit images without a clear understanding of what's it all about and I should say many armchair experts posting here are clueless of the differences as well.

Essentially, there are 3 luminosity channels (RGB) in every image captured by the sensor. In an 8 bit jpeg image, there are 256 shades of red, 256 shades of blue, and 256 shades of green. That's 8 bits per color, or 256 x 256 x 256 or 16,777,216 possible colors. A 12 bit image records 4096 shades per channel while a 14 bit image records 16,384 shades of color for each channel. So it is pretty clear that a 14 bit image records (4.3 trillion possible colors) 4 times more data per channel than a 12 bit image (68.7 billion possible colors), and a 12 bit RAW file has 16 times more data per channel than an 8 bit jepg image. If you only shoot jpeg, then all this is pointless from hereon. If you want every bit of image data that the sensor can deliver, you've got to shoot in RAW. And to answer your question, the file size for a 14 bit image is going to be a lot bigger than a 12 bit image, which has a bearing on speed of camera operation.

However in practice, there are limitations on the color range (gamut) your monitor or printer can display. So while there's a definite difference between a 12 bit and 14 bit file RAW image that the camera captures, most here will probably never have equipment that will be able to display the difference, nor use software that can handle 14 bit files or post process extensively to be able to exploit the difference.

[madbrain]

I'm glad to hear that. It is encouraging. Let's hope then, that the K-30 is much closer to the K-5 than the K-01.
Somewhere in Frank's long thread, he provided links to the actual RAW files (which you could download). Those links are probably dead by now. I forgot about that fact until you pointed out that those are JPEGs to which I linked. Anyway, I noticed the difference in quality in the actual RAW files too.

OK. I didn't see the RAW files. But looking at the JPEGs, it was obvious to me that the colors were slightly different. This could easily be due to differences in the JPEG engine.
RAW files would tell a bit more because there is less processing. In particular, they are losslessly compressed, unlike JPEGs.
There could still be other differences in the cameras, and not just the 2 extra bits.

My main point is that you still cannot see the extra color bits on your display. Most monitors and video cards still only display 8 bits per channel (R, G, B), ie. 24 bit colors total. It doesn't matter if your files have 12 or 14 bits per channel, they have to be rendered as 8 bits in the end.

To give you an analogy with audio recording which I also do in my pastime, professionals use 24 bit for recording. It doesn't actually sound better than 16 or 20 bit recording by itself. However, 24 bit audio allows for mixing multiple channels and doing some editing and effects, without compromising the overall quality. The audio is only downconverted to 16 bit at the end of all the processing steps, for example when producing a CD which uses 16-bit audio.

IMO, the story is somewhat similar with images. If you think about it, 12 bits per RGB channel is already 36 bits. That is 64 billion individual colors. Even if there was a display that could display that many, I am not really sure that your eyes could differentiate them. If you have 14 bits per channel, then that's 42 bits total or 4.3 trillion different colors. Again, I am very doubtful your eyes could tell that many colors apart.

If you don't do post-processing, I don't think anything beyond 8 bits matter. I have shot in JPEG the past 4 years with my K200D and now K-r. I checked my hard drive and there wasn't a single PEF or DNG file that I kept long-term. Indeed, I wasn't doing any post-processing. Now that I am starting to do some, I am starting to shoot in RAW a little bit . In the end though, your 36 bit or 42 bit RAW files are probably going to be downconverted to some 24 bit file format, and even if not, they will be downconverted to a 24 bit display for sure. Perhaps I will find there are some old shots I would like to PP and wish I had RAW files for. That hasn't been the case so far.

BTW, I use several large displays on my home PC, two HP 30" with 2560x1600 resolution, and a third at 1920x1200 in portrait mode. They are all 24 bits color. I don't see any color banding. It's not obvious what the benefit would be of having a 36 bit or 42 color monitor and video card. I would much rather have more MP. Going by 900,000 pixels on a 3" LCD on a typical camera these days, I should be able to have 90 MP on one 30" LCD using the same pixel density

Oh, and I'm a software engineer if any of this matters. I have written audio mixing software and graphics code in my teens for pay before. That was a long time ago. Not so much these days, I do networking and security work.

Nonetheless, I sincerely hope you are right about the comparison being between two cameras and not between 12 bits and 14 bits.

The entire thread is about K-01 vs K5, I don't see how there can be any doubt on this matter.

[madbrain]

There are people here on PentaxForums and elsehwere (including actual reviewers) who say (or at least imply) that those extra two bits make a meaningful difference; however, since I have never shot in RAW, I can only take their word for it and your word for it. For my sake, I sincerely hope you are the one who is right.

The 2 extra bits can only make a difference if you shoot in RAW, and then only if you do post-processing. If you don't do either of those things, they don't really matter.

However, I must ask: If the differences are really as negligible as you say they are, why does Pentax treat 14-bit RAW as if it is only for higher-end pros? Why don't they also give it to the slightly more consumer-oriented K-30 and the K-01?

Speed could be one reason, you need faster processing for the higher 14 bits data rate . You may need a faster SD card too. Even if the camera doesn't record all 14 bits, there may still be processing done internally on 14 vs 12 bits before the files are eventually downconverted to 8-bit JPEGs. That is most likely the case on the K-5.

Is the 14-bit-vs-12-bit debate being used as a cheap, deceptive marking gimmick?

As I said before, I think it's a bit silly to cripple the camera down to 12 bits if all the hardware can actually do 14 .

But yes, this is what marketing people do sometimes.
My father told me a story of how IBM once "upgraded" their 10MB hard drive to a 20MB . This was a very expensive upgrade. They sent a tech to do it. They didn't swap any hardware. The tech just changed a jumper on the drive...
If the hardware is capable of 14 bits then some hacked firmware could presumably unlock it.

I wouldn't be surprised if it is. And it also cannot be argued that Pentax is taking consumers' and enthusiasts's hard drives into consideration, because most of them these days have just as much hard-drive space as the pros do, so file size cannot be used as an excuse. I, as a Mac technician, certainly have tons of hard-drive space at my disposal.

I agree hard drive space is not that much of an issue these days for consumers. However once you start adding the cost of backups, buying extra drives adds up. I probably have over 20 TB worth of hard drives worth just for backup purposes for 6 different computers...

[Welfl]

Welfi, like Pentaxor, I think you're overthinking and fussing about the differences between 12 and 14 bit images without a clear understanding of what's it all about...

Yes, I admitted earlier that I was probably "fussing" a bit more than was necessary.

and I should say many armchair experts posting here are clueless of the differences as well.

Ah, yes, but I don't own an "Armchair-Expert Detector" ; therefore, for the sake of being kind and courteous and not wanting to start a real argument and/or not wanting to sound rude, I have intentionally chosen to treat everyone as if he or she is a real expert, until I learn otherwise, and even then I will continue to remain polite.

Essentially, there are 3 luminosity channels (RGB) in every image captured by the sensor. In an 8 bit jpeg image, there are 256 shades of red, 256 shades of blue, and 256 shades of green. ...

Yes, I learned about this last night when I was doing some research over at SnapSort.com. To save time, and to avoid repeating myself, I will explain this in greater detail in my reply to "madbrain."

So it is pretty clear that a 14 bit image records (4.3 trillion possible colors) 4 times more data per channel than a 12 bit image (68.7 billion possible colors), and a 12 bit RAW file has 16 times more data per channel than an 8 bit jepg image.

Leaving aside the part about colors, what you have written is basically what I have read in the past about 14-bit RAW files (i.e. the additional data), but I must once again refer you to my second sentence above.

If you only shoot jpeg, then all this is pointless from hereon. If you want every bit of image data that the sensor can deliver, you've got to shoot in RAW.

I do not intend to shoot jpeg anymore (with occasional exceptions, of course) because I know the benefits of shooting RAW. I've been a Photoshop post processor for many years with scanned images. I love making them look better than they ever looked before.

And to answer your question, the file size for a 14 bit image is going to be a lot bigger than a 12 bit image, which has a bearing on speed of camera operation.

Thanks for explaining that. It clarifies a lot for me. Madbrain also gives the same explanation.

However in practice, there are limitations on the color range (gamut) your monitor or printer can display. So while there's a definite difference between a 12 bit and 14 bit file RAW image that the camera captures, most here will probably never have equipment that will be able to display the difference, nor use software that can handle 14 bit files or post process extensively to be able to exploit the difference.

Yes, in all of my discussions about 14-bit RAW vs 12-bit RAW, color was not what I was thinking of, because other people over the past year and a half didn't explain themselves very well and because I didn't understand. But I think there are still some visible benefits in spite of the number of colors a computer screen can show. I believe the principle may be similar to what I gain when I improve scanned images in Photoshop using the "Selective Colors" menu. To save time, I will explain in greater detail in my reply to madbrain.

Thank you very much for your well written and polite reply.