Wednesday, January 6, 2010

PC's and Apple allows third party software, Why can't Pentax do that?

The above is what writing code for Pentax Firmware looks like.

Visitors to this page also liked: See "Recommended Reading" on the right column.We honor reciprocal links. Email: brqyvn@aol.com

Hi Pentaxian friends.

We all know that digital cameras are somewhat like computers. Sure, they use real glass lenses and many of the traditional analog cameras characteristics are still utilized. However, some of the functions , if not most of them, are embedded into the camera's "Firmware". Currently, only Pentax can modify the Firmware and they do so by issuing downloadable revisions mostly to fix problems reported by users.

You know that the cheaper DSLRs could have some of the same functions as the more expensive models. For example, the new K-x does not show the "red-illuminated" square over the focus point. I suspect that it could easily be added in the Firmware. The K20D and the K-7 have the capability of adjusting the front/back focus of any lens attached. Again, I suspect that this feature could be added in the Firmware. I remember the complaints from K10D owners wanting to have the K10D updated so that it too would have the capability of adjusting front/back focusing. I also remember some hacker, from Russia, coming up with some solution to the problem, that anyone could somehow enter in the firmware. Of course, any problems caused by someone unsuccessfully trying to modify the Firmware, automatically voided the warranty.

There are many great software companies in the world, that could come up with pretty interesting features. Why not grant a license to these guys and let them sell additional features for the Pentax cameras. Instead of reducing the sales, I believe it would increase them.

Would you buy an Apple computer if only Apple were able to write software for its computers? Would you buy a PC if it would only run on the manufacturer's software? Of course you wouldn't.

Pentax has always been a unique company with design and ideas outside the box. Maybe time has come. The time to beat Canikon at the game has come...maybe.

What do you think?

Thank you for reading,

Yvon Bourque

15 comments:

s e a n said...

If I am not mistaken the K-X does not have that the red dot focus indicator because the camera itself had that feature (red leds overlaying the viewfinder) physically omitted to have a smaller camera body, so it is not a software issue.

admittedly, i have become dependent on the focus indicator, especially in low light scenes, and this (if a little nitpicky) and the lack of battery grip keeps me from getting the K-X. On the other hand, it has that wonderful sensor, great ergonomics (relative to its size) and great price tag.

JohnP said...

I think it would be a great idea as far as users go but I can't see any camera maufacturer allowing it. They like to sell a range of cameras from cheap to pro.Allowing the cheaper models (that quite often share the same sensor if not all the same other hardware) to be updated to near pro via 3rd party firmware would damage their marketing strategy. Why buy a pro spec model if you can buy a cheapie and update it? Unfortunately all manufactures leave out software features in lower end models to encourage us to lash out more for the more featured models.

s e a n said...

good point JohnP, but there are limitations to how much software/firmware can improve a camera's performance.

What are the main differences between low-end and pro models?

generally pro models have weather sealing, more sophisticated AF systems, imaging engines, bigger sensors, integrated battery grips, better shutter actuators etc.

more often than not the things that separate low-end from pro models are physical features which software/firmware cannot replicate.

Anonymous said...

Inspired by the recent post of Thom Hogan? (bythom.com)

Anyway this is so obvious to me that I can't understand no company has done it yet. So many people/fanboys would be willing to improve their cameras that we would probably see huge improvements in the mid-term. Or we could fix annoying features that we don't want (e.g. the DFS in k20d/k-7, which I cannot be disabled on the contrary to the k10d).

Michael said...

One of the major problems with hacking DSLR firmware is not a technical one, but more a social barrier. There are just too few people round there that are a) fotografers and b) programmers. The illustration in your article is just plain old assembly code, but these days programmers with knowledge in machine language are rare, even rarer for those Image-Processors companies use.
There is a pentax hackers site out there (the link is on my home pc, so I can’t post here :-) but they made promising progress first, seem to be stalled right now.
Pentax has an even smaller user base then other brands so alas I, too, hope for a pentax hackers devolper kit (phd-k) for my K-m that is not going to happen. Sigh.

tmt said...

The site mentioned is http://www.pentax-hack.info

There are some really good info to the current Pentax camera line, the main difficulty is that you can reverse-engineer the assembly code but it is very tough stuff. Programmers (like me) nowadays used to higher level programming languages, but it is not possible to exactly translate back the assembly code to C or C++ or whatever was the language developers used.

Michael said...

One thing the pentax-hack guys suffer from is the fact, that there is no documentation on the internal registers and adresses used. Thats the major problem in reverse-engeneering code and takes up the most time: to figure out the mapping to apply. So if any pentax engineer involved or the guy at the OS manufacturer would risk their working place and give them some hints …
Come on guys, just for the good of open source community!
The best thing would be the back-contributions that usually occur from such projects. Great ideas, that Pentax could take up and put into their regular models.
I don’t think that the economic reasons would be a big thing. Look at Canons compacts and the chdk. Did CHDK affect canon sales numbers in any way? Nope, it just contributed to some happy users wich would otherwise have changed the brand for a more elaborated one maybe.

Miserere said...

I suggested something similar in an article I wrote back in May 2009. My approach, however, was to leave the firmware alone, and have camera companies allow the addition of software modules written by users or 3rd parties.

Camera companies would also get on the bandwagon by releasing improved firmware for purchase. In this way they could afford to release fewer bodies because they would be releasing firmware upgrades.

We are at the point where DSLRs are not really improving that much, physically. Many improvements are coming from the software side. For example, the K2000 used the same 10MP sensor as the K10D, yet could shoot at ISO3200 while the K10D only reached ISO1600. I would pay $100-200 for an upgraded firmware for my K10D that would provide me with ISO3200 capabilities. They could also include an AF adjust setting similar to that introduced in the K20D. The K10D has a smaller memory so it would only allow for 4 lenses (and a global adjustment); this was confirmed to me by a Pentax technician.

I don't understand why they don't sell upgraded firmware for older cameras; the benefit margin is so much higher than that of a camera because you buy directly from the camera company's website and it's a simple download.

Kym said...

Its not a Pentax issue but rather all camera companies.
The real issue is what is known in IT as free open source software.
It's what Linux, The GIMP (as some examples) are all about.
Microsoft don't lets us touch their code. They only let you load other programs on top of Windows. You cannot (or are not supposed to) modify Windows components.
An extensible firmware would be cool but for $ and support reasons it won't happen.

Yvon Bourque said...

I'm obviously not a software engineer, but why not engineer a way to use an additional card slot (SDHC or whatever)and get additional functions by inserting the card. Pentax wouldn't have to give the codes to anyone. They could sell the additional functions already loaded on the card. It has to be cheaper to make software than cameras. I suspect the markup would be many times that of building cameras. It just makes sense to me.

parkij@ihug.co.nz said...

What a great idea! No doubt it will get lots of support. This could be the making of Pentax. The big problem is verifying the suitability of any mods suggested.It could be quite simple. You have a mod. Pentax checks it (say for $10) and if its Ok markets it for $20 as a download. Pentax can then become a customised specialised camera you can customise. WOW! Big money to be made and lots of happy customers
parkij

JohnP said...

Sounds feasable if their is money in it for Pentax. We can only hope.
Admittably not Pentax but another manufacturer's employee has told me they actually spend money to downgrade software used in high-end models when introducing cheaper models. It's part of their market saturation policy.

Unknown said...

the only way to get hackers interested in pentax is by putting video games in the cameras. I just saw a big documentary how there is such an underground society that hacks sony psp's. Sony can't fight them. No sooner do they do something to fix the issue that the hackers fix it.

As for pentax their writing is copyrighted. I'm sure Nikon canon or sony wouldn't mind taking a peek at how their processing engine codes and decodes to get such awesome images. They had three cameras with the k10 sensor i'm sure the kx is not the last camera they will have with the new sony sensor. maybe a k300 or a k30 sometimes soon, but also i've hear and that full frame is coming soon.

Georg said...

Hello Yvon,

At first I was embarrased on hearing that point from you.

But then I found , that this could be a perfect idea, to improve the cameras features.

So ; i can second your idea, to encourage PENTAX, to go such an unusual way .

Best regards from ´Germany
Georg

Unknown said...

The comparison with computer operating systems is interesting. There are, to date, no open source operating system that has delivered a comprehensive and successful version for the end-user desktop. With emphasis on successful. Linux has a lot of good things going for it, but it continuously fail to appeal to the masses. Why would anyone believe that camera firmware is any different?

By the way, it would be more correct to compare the firmware of cameras with BIOS firmware for PCs. The BIOS can be upgraded, but follows the same hazard and clunky procedure as do camera firmware upgrades. Moreover, most PC/motherboard vendors have "warranty void" exclusions for end users attempting upgrades on their own. Just like the legal text attached to camera firmware upgrades.

With these similarities in mind, It's interesting to note that the specifications for INTEL's chipsets are open enough for some reputed BIOS makers (AMI, Phoenix) to make third-party firmwares for motherboard makers like ASUS and MSI. But where are all the opensource BIOSes for INTEL chipsets? Maybe they're nonexistent for a good reason?

An example more in the "devices" category is wireless network routers. One vendor, Linksys, opened the source for one of their routers. A move that was much hailed in the opensource circles. The only outcome, however, was a bootleg firmware version violating the regulations on radio signal wattage to improve the range of the unit. Linksys stopped distributing their source when it was clear that the only contribution from the community was to break laws.

I do not imply, by the last example, that independent makers of camera firmware will break any laws. However I do believe that they will emphasise certain "features" tailored to personal desires. Like developing with blinders. A worst-case scenario is that some implemented features screw up others in a bad way. I've seen enough freeware doing just that with my computers, to the point of becoming very weary of the concept all together. It's a sad fact that most open source development projects never get much further than the proof-of-concept stage.

Sorry to be such a party pooper.