[TomGarn]

Yes, Mark ...
That is right Imageman ...

As said above this is what I did - me, the poor guy with a dead second hand K5 for 350,- € - when I bought another used K5 for 350,- €:
I asked the sellar to do the newest firmware update before our financial transaction took place. He did so - with success.

I would only do another (unexpected to come) firmware on this K5 - if it corrects the stupid > ISO-Shutter-Automatic < in video mode
But if they really worked on this, the K5 would be a much to good deal for video-makers with small budget and the wish for a good codec and SR.
But they don't seem to care to much for video - even with the K3 (which got a worse codec and a bad SR - but of course better "sound and light")

Watch out for your fresh K3 Update now ... maybe invite some witnesses before updating - just in case of ... ?

---

Last edited by TomGarn; 07-13-2014 at 02:59 AM.

[MarkJerling]

My K-5 is still under warrantee, but having said that, I have performed every firmware update without any problems. I use an old, but good 64mb SD card, which I use only for that purpose, and I always make sure I insert a fully charged (factory not aftermarket) battery. But, I agree, a firmware update should not be able to brick a camera, or any other expensive equipment for that matter.

[VisualDarkness]

All in all they should have a backup boot system so that in case of failure you would be able to apply a new firmware. With the memory costs today it wouldn't cost much.

[clackers]

All in all they should have a backup boot system so that in case of failure you would be able to apply a new firmware. With the memory costs today it wouldn't cost much.

The dual-BIOS motherboards served this function on PCs.

A problem with cameras is that SD cards are quite variable in their quality, and a bad read equals failed update.

[TomGarn]

Maybe in the future this old problem about firmware-updates is solved, and there can be a safe and secure chance to do it again ...
But because it happens so rarely the engineers have not given a second thought. Or is it to expensive ?

The root cause for sudden death (if not an empty battery) can be a corrupted file because of bad unzipping or a hidden problem
of the memory card ... or the hardware got a problem you may not have been aware about yet ...

I want to ask again, how many cases of firmware-update-failures we can document here:
Here I recount 3 (almost 4) cases:

- - - - Dead K-7:
Electric Eye (TO)
Ex Finn.

- - - - Dead K-5
TomGarn

++++ Luckily repaired failure on K-5 and Q
p38arover

[Imageman]

The thing is people assume that a manufacturer has the interests of the customer at heart. The sad truth is most managers in manufacturers like to pretend they do but they are only interested in looking good to their boss so they get a bonus and the next promotion. Looking good means not wasting the companies money helping customers, unless forced to.


The customers problems will be attended to under warranty, and if the unit is outside warranty then who cares. If it breaks down let the customer pay for the repair or buy a replacement.


They don't care if its a common problem as long as the customer doesn't complain about it and make extra work for them fixing it. If the customer wants to pay for a repair that's just fine of course.


If it is a common problem and it will cost lets say 100,000 dollars to fix over the life of the camera model, but it will cost them 90,000 dollars in warranty repairs if they don't bother fixing it, they wont bother fixing it. And nobody is going to know if its a common problem or not if everyone in Ricoh keeps hiding the issue.


Look at whats happening, A camera that costs 800 dollars is unlikely to have cost more than 200 dollars to build, the rest is profit and additional costs such as advertising packaging and shipment. The firmware and associated memory chips would cost around 10 dollars of that or less, so its quite possible that a long term fix would cost Ricoh an extra 5 dollars on the camera cost to implement. Its easier or them to make the customer pay 700 dollars to fix a bricked camera, than to put the fix in the camera at the factory for 5 dollars as long as the customer doesn't make waves but instead pays his 700 dollars to fix a problem that is probably Ricohs fault anyway.


Either nobody is affected, or nobody in Ricoh cares about it, or nobody has complained enough.

[FrankC]

Maybe in the future this old problem about firmware-updates is solved, and there can be a safe and secure chance to do it again ...

Actually, the technology is already here. Current mid to high-end motherboards have the ability to store the current BIOS version during an update. If the BIOS flash process fails, then mobo can have its BIOS reverted back by the user. However, this can only be done by pressing a button on the motherboard itself and not by pressing a button on the keyboard during the boot process. Nonetheless, I don't see a reason why there can't be a dedicated button or set of buttons that can be pressed during the boot up process that reverts back to the backup firmware.

[Imageman]

Ive argued several times in different threads that managers in businesses don't care about customers or doing the right thing when making business decisions and also that they use cost benefit analysis to determine the most cost effective course of action.


This means that it doesn't matter if a solution exists that is easily implemented, they simply add up the cost of implementing it multiplied by the number of units they expect to sell and then see if theres a good financial reason to spend the money. In other words they will only make the change if they either make greater savings by putting it in or will make more profit.


I have worked in business and been a part of this process, but I think many are still unconvinced, and that's ok it can be a difficult pill to swallow. So id like to relate the Ford Pinto experience, which is well documented and is a good example of how manufacturers operate in this scenario of justifying an unexpected change to a product.


Remember this actually happened and everything in this excerpt is true and a matter of record.


It was the late 60s, when the demand for sub-compacts was rising on the market. The specifications for the design of the Pinto were uncompromising: "The Pinto was not to weigh an ounce over 2,000 pounds and not cost a cent over $2,000." During design and production, however, crash tests revealed a serious defect in the gas tank. In crashes over 25 miles per hour, the gas tank always ruptured. To correct it would have required changing and strengthening the design. A journalist covering the story wrote:

"When it was discovered the gas tank was unsafe, did anyone go to Iacocca (the project manager) and tell him? "Hell no," replied an engineer who worked on the Pinto, a high company official for many years… "That person would have been fired. Safety wasn't a popular subject around Ford in those days. With Lee it was taboo. Whenever a problem was raised that meant a delay on the Pinto, Lee would chomp his cigar, look out the window and say, 'Read the product objectives and get back to work.'"

In 1965, Ralph Nader had brought automobile safety to the public's attention with his book Unsafe at Any Speed. Government was just beginning to regulate automobile safety in those days, but Ford had a way of getting around it. Lobbyists for Ford and other auto-makers convinced the government to delay regulations on fuel tanks for eight years.

One of the tools that Ford used to argue for the delay was a "cost-benefit analysis" of altering the fuel tanks. According to Ford's estimates, the unsafe tanks would cause 180 burn deaths, 180 serious burn injuries, and 2,100 burned vehicles each year. It calculated that it would have to pay $200,000 per death, $67,000 per injury, and $700 per vehicle, for a total of $49.5 million. However, the cost of saving lives and injuries ran even higher: alterations would cost $11 per car or truck, which added up to $137 million per year. Essentially, Ford argued before the government that it would be cheaper just to let their customers burn!

Of course, the public eventually learned that the Pinto had a tendency to explode in rear-end collisions, and victims and their families sued the company. Jurors were outraged over Ford's low value of human life and awarded the victims huge settlements. However, the final shocker came when Ford actually got around to fixing the flawed gas tanks. It turns out that the "cost-benefit analysis" that Ford submitted to the government was entirely bogus: the cost of fixing each car was not $11, but merely one dollar.


So Ford Motor Company refused to choose customer safety over cost, and used projected costs of a fix in comparing the estimated costs they faced by doing nothing, doing nothing won because it cost less to sit around pretending that there wasn't a problem and letting their customers die horrifically in unnecessary fireballs trapped in their cars. And the figures they based these vile calculations upon were fabrications to make themselves look good and fool the government into doing what they wanted.


Can anybody believe that companies base decisions on anything other than cost and benefit to them. I hope not.

[VisualDarkness]

Fantastic story and as an economics student I know exactly what you talk about.

[TomGarn]

Not more than 3 people here who have reason to really complain ? (like me)
Then this problem rather is a question of garanty and policy ex gratia (Kulanz)
And a question about a good straight honest chat with your friendly Pentax/Ricoh-Dealer.
(For I have reason to may be lucky eventually soon ...)

I want to ask again: How many cases of firmware-update-failures we can document here ?
Here - in this thread - I counted 3 (almost 4) cases:

- - - - Dead K-7:
Electric Eye (TO)
Ex Finn.

- - - - Dead K-5
TomGarn

++++ Luckily repaired failure on K-5 and Q
p38arover

[magkelly]

If it's not broke, don't fix it. That's my motto when it comes to firmware and it has been ever since my first tablet bricked after an update to it's firmware. That was a cheap $50 used tablet and it ticked me off. Now imagine if that was a camera worth $500 plus. My sexy red K-30? I'd be totally irate and I'd let Pentax-Ricoh know it too. I know there are probably things in the updates that would probably benefit the camera, but I'm still very leery of going there with a device that costs that much, and one that I can't just reboot and reload if an update fails. Particularly since I don't have a solid year warranty on the K-30, it being a used camera and all.

[TomGarn]

These two threads now have got 13.700 views - and only 3 dead cameras listed yet.
So be realistic please - take it easy - relax ... Or add your case to the lists:

Users in this Forum with sudden death of their Pentax
after corrupted firmware-update are:

2 x dead K-7:
Electric Eye
Ex Finn

1 x dead K-5:
TomGarn

2 x repaired failures on K-5 and a Q
p38arover

Read more at:
https://www.pentaxforums.com/forums/115-pentax-k-5/253883-new-k-5-firmware-ve...#ixzz39hCsJkyv

[VisualDarkness]

It has almost gone into some kind of hysteria here in the thread, nothing points to high risks.

[robthebloke]

I'm interested how on earth does firmware upgrade cook a main board?

Kinda makes upgrading a possible scary option.

Firmware for electrical devices is usually stored in a surface mounted EEPROM chips. You could (theoretically) remove the chip, flash it with the firmware, and then resolder it to the board; but that's not exactly an easy thing to do without specialised equipment. (It's unlikely a camera repair shop would have those tools).

If pentax included a header on the board to allow them flash the rom in a service centre, then it would be fixable.

Chances are they don't have that, and they most likely pre-flash the rom chips prior to assembly (because it's easier/cheaper).

If the update fails, you'll no longer have as OS to be able to bootstrap the SD card for the rom update. The components will all be fine, but the board would need a factory refurb to get the OS booting again. Replacing the board is the easiest option (just requires a screwdriver).

---------- Post added 08-08-14 at 01:31 AM ----------

Ive argued several times in different threads that managers in businesses don't care about customers or doing the right thing when making business decisions and also that they use cost benefit analysis to determine the most cost effective course of action.

Sure, pentax could build in a second EEPROM. They could wire it up to all components on the board. They could add additional logic to make it switchable.

The problem is that all of this stuff adds complexity, draws more power, needs larger circuit boards, adds more points of failure, and has more downsides than benefits to the majority of consumers.

Do you want a camera the size of a canon, with worse battery life, a greater chance of failure, and a higher price tag?

Or will you make sure you format you sd card prior to a firmware upgrade?

---------- Post added 08-08-14 at 01:51 AM ----------

You'd think that Ricoh/Pentax would provide an MD5 hash on their download site so folks could check file integrity after a download.

All pentax firmware already includes a checksum. If you try to load a corrupted firmware onto the camera, it will refuse to load it - see the work that Shodan has done to reverse engineer the firmware checksum here: https://www.pentaxforums.com/forums/6-pentax-dslr-discussion/250555-resurrect...hacking-5.html

Charge your battery fully, and format your SD card before flashing the firmware. Pentax safeguards against corrupted firmware, and will even tell you that your SD cards have issues if you format you SD card first.

[Imageman]

Sure, pentax could build in a second EEPROM. They could wire it up to all components on the board. They could add additional logic to make it switchable.

The problem is that all of this stuff adds complexity, draws more power, needs larger circuit boards, adds more points of failure, and has more downsides than benefits to the majority of consumers.

Do you want a camera the size of a canon, with worse battery life, a greater chance of failure, and a higher price tag?

Or will you make sure you format you sd card prior to a firmware upgrade?

That's a ridiculous to say, people who formatted their sd card have still suffered corrupted firmware.


The fact is managers only concern themselves with cost in their decision making and it will always be so.


And as for your comments suggesting a fix would increase the size and complexity of the camera, that's not true all they would have to do is wire an external port and then modify the firmware so that the camera looks for firmware first in the eeprom and then if that fails, it looks for a firmware image attached to the external port, and if it finds firmware there it loads it from there.


Repair centres could then when presented with a bricked camera, attach an external device with firmware to this port, and reload the firmware to the eeprom after successful camera boot.


no extra eeprom no extra weight no extra battery useage, negligeable extra cost.


Wheres the issue.