jaapv Posted December 14, 2009 Share #81 Posted December 14, 2009 Advertisement (gone after registration) Frank you suggesting software should not be tested, bugs taken out and retested? That sounds like bollocks to me. Link to post Share on other sites More sharing options...
Advertisement Posted December 14, 2009 Posted December 14, 2009 Hi jaapv, Take a look here Firmware Update -- Predictions on When??. I'm sure you'll find what you were looking for!
Guest BigSplash Posted December 14, 2009 Share #82 Posted December 14, 2009 No Frank. Legally, I'm totally correct. Leica does not sell directly to the public, therefore it has no legal contract with any end user. Fitness for purpose is a matter between the retailer and their customer. Surely as a CEO you should know that. Maybe this will explain where Leica's focus is: Leica Camera AG - Corporate Values Redbaron how can you say this? Leica does not sell directly to the public..... Wrong! Mayfair and the other Leica Boutiques are fully owned Leica subsids. tat do sell to the public Manufacturers have a duty to provide product that is fit for purpose to their dealers, distibutors etc....When I worked at Texas Instruments we settled a massive (1B$ write off) for the home computer that started as a result of power transformers catching fire. The faulty tranformer was bought from a subcontractor that was supposed to be UL approved and was not. The responsibility was with TI although the item was sold via distributors and then retailers. Anyhow I repeat that Leica will never in my view duck their legal responsibilities as implied above. They will try to bring out firmware improvements. Link to post Share on other sites More sharing options...
stunsworth Posted December 14, 2009 Share #83 Posted December 14, 2009 Andy is that a guess that Leica have had 4 iterations since September? Frank, there's a clue in "there may well have been". Do you ever bother to read Andy's posts before responding to them? If it is true they must be taking a scattergun approach to what should be rigorous software design using a solid configuration management tool to ensure that what they change does not affect other parts of the software. I don't understand what you mean Frank. Imagine the following scenario. You make an alteration to function A. You then do a regression test to make sure it doesn't affect anything else - always a possibility with software. You then amend function B. This too is followed by another regression test. You continue until all the desired functionality is included. The alternative - which you appear to be proposing - is that all the changes are made in one 'big bang' following which the software engineers have to try and work out if any bugs are caused by the new modules interacting with each other or with existing code. Link to post Share on other sites More sharing options...
Guest BigSplash Posted December 14, 2009 Share #84 Posted December 14, 2009 Frank you suggesting software should not be tested, bugs taken out and retested? That sounds like bollocks to me. Jaapv I am not saying that the software should not be tested but I am saying that your methodology is wrong. Clearly you have no idea of software devt....I have many years of experience in this field! I guess you should stick to dentistry. FYI The engineers will need to identify the issue that they are trying to resolve and associate that with a paticular program routine, embedded in the firmware. They obviously will need to fix the bug (eg ISO number) in that routine. However invariably this will cause consequent issues or new bugs in one (or possibly many more) other routines (eg shutter speed, EV correction, lens corner issue - vigneting etc etc) which the engineers need to consider. To achieve an understanding of that they use a configuration management tool to see what routine will affect other parts of the software. If the architecture is none modular it obviously gets much more difficult, and time consuming....but none modular usually means more performance within a spaghetti set of code. The suggestion that they are testing , remove a bug, and then retest such that they have had 4 firmware releases since Sept if true is not a rigorous approach and would indeed be bollocks if Leica are doing that, which I doubt Worse such an approach may even create a firmware that if released shows up issues in wierd situations down the road...Certainly such an approach will be difficult to use as a basis for the subsequent releases of firmware. A rigorous design methodology is the only way to avoid such a situation and Leica has had a considerable period of time to do the job properly. We are now in December. Link to post Share on other sites More sharing options...
Guest BigSplash Posted December 14, 2009 Share #85 Posted December 14, 2009 Frank, there's a clue in "there may well have been". Do you ever bother to read Andy's posts before responding to them? I don't understand what you mean Frank. Imagine the following scenario. You make an alteration to function A. You then do a regression test to make sure it doesn't affect anything else - always a possibility with software. You then amend function B. This too is followed by another regression test. You continue until all the desired functionality is included. The alternative - which you appear to be proposing - is that all the changes are made in one 'big bang' following which the software engineers have to try and work out if any bugs are caused by the new modules interacting with each other or with existing code. Steve see my reply to Jaapv...obvioulsy you do regression testing...I mention this. Are you saying that they should not use a config. management tool and just keep bashing away until it works at test? If someone says "may well have been released 4 times" why not 2, 6,8, 20 times? I tend to believe that the person has knowledge and is not just ........... ! Link to post Share on other sites More sharing options...
jaapv Posted December 14, 2009 Share #86 Posted December 14, 2009 Jaapv I am not saying that the software should not be tested but I am saying that your methodology is wrong. Clearly you have no idea of software devt....I have many years of experience in this field! I guess you should stick to dentistry. Like you stick to camera designing ? I suggest you actually read Steve's reply, You may realize that you made a 180 degree within three posts.... Link to post Share on other sites More sharing options...
chris_tribble Posted December 14, 2009 Share #87 Posted December 14, 2009 Advertisement (gone after registration) Steve - seriously - the IGNORE button is so useful. There's none so blind as those who cannot see... Link to post Share on other sites More sharing options...
jonoslack Posted December 14, 2009 Share #88 Posted December 14, 2009 If someone says "may well have been released 4 times" why not 2, 6,8, 20 times? I tend to believe that the person has knowledge and is not just ........... ! Well, surely 'released' is the operative word here - if there were 4 releases of the beta firmware, that might actually represent hundreds of different builds, with a proper routine of internal testing and a configuration manager. As for believing that the person has knowledge . . .when they specifically say that they don't have knowledge! It seems to me Frank that without you go spend a few weeks seeing what Leica actually do, you are simply firing blanks off into the darkness. Link to post Share on other sites More sharing options...
Guest BigSplash Posted December 14, 2009 Share #89 Posted December 14, 2009 Like you stick to camera designing ? I suggest you actually read Steve's reply, You may realize that you made a 180 degree within three posts.... Jappv please look again at my reply to you and the one to Steve. I have NOT changed anything within the three posts....please show me where there is ANY inconsistency. Frankly I give up ...time is really too short... The great thing about this thread is that I have found somewhere to go and get a Trielmar lens hood and for that I appreciate being involved with the forum. Link to post Share on other sites More sharing options...
stunsworth Posted December 14, 2009 Share #90 Posted December 14, 2009 Steve - seriously - the IGNORE button is so useful. There's none so blind as those who cannot see... I know, but where's the fun in that? The people who you'd want to ignore are invariably the ones that give you the biggest chuckles at other times. Frank's bender about Leicas firmware testing methodology is based on Andy Barton saying they may have had 4 iterations of firmware in the pipeline since the M9 was launched. Since what Andy said was speculation, everything that Frank has written since it is pointless. He doesn't know, I don't know, no one outside of Leica and possibly a small group of beta testers know how they develop the firmware. In the past Leica have released firmware updates in beta to selected individuals. Whether any one beta becomes the release candidate will depend partly at least on the results of the beta test. Again, Frank - despite being a former CEO of an electronics multinational - knows no more about how this is preceding that any of us here (beta testers excluded). Link to post Share on other sites More sharing options...
Guest BigSplash Posted December 14, 2009 Share #91 Posted December 14, 2009 Well, surely 'released' is the operative word here - if there were 4 releases of the beta firmware, that might actually represent hundreds of different builds, with a proper routine of internal testing and a configuration manager. As for believing that the person has knowledge . . .when they specifically say that they don't have knowledge! It seems to me Frank that without you go spend a few weeks seeing what Leica actually do, you are simply firing blanks off into the darkness. I agree.with what you say about releases and builds. I also agree about shooting off blanks into the darkness.... My comments were made in good faith and were based on inputs posted here that I believed to have some credibility. It is also based on managing software design teams. If I am in fact commenting on speculation, and ill informed opinion that is in any case actually wrong ...... It is not useful at all I agree. Link to post Share on other sites More sharing options...
Nicoleica Posted December 14, 2009 Share #92 Posted December 14, 2009 Excuse me for butting in here. but I believe that the magic number 4 comes from an earlier post which stated that the author had handled a camera with version v1.006 firmware installed. As the current public release version is v1.002, one could reasonably expect that this was 4 versions later. However, in the software world at least, actual releases do not necessarily have to follow this rule, and the next public release could v1.100 or even v2.000, depending upon how significant the changes are deemed to be by the vendor. Link to post Share on other sites More sharing options...
andybarton Posted December 14, 2009 Share #93 Posted December 14, 2009 Frank's bender about Leicas firmware testing methodology is based on Andy Barton saying they may have had 4 iterations of firmware in the pipeline since the M9 was launched. Since what Andy said was speculation, everything that Frank has written since it is pointless. He doesn't know, I don't know, no one outside of Leica and possibly a small group of beta testers know how they develop the firmware. My "speculation" is based on the fact that M9s ship with v 1.002. A member has held an M9 with v 1.006 That looks like 4 iterations to me. OTOH, it could be 400 iterations, so yes, mea culpa. I am ill-informed. My New Year's Resolution is to stop trying to talk sense to brick walls. Link to post Share on other sites More sharing options...
jonoslack Posted December 14, 2009 Share #94 Posted December 14, 2009 My "speculation" is based on the fact that M9s ship with v 1.002. A member has held an M9 with v 1.006 That looks like 4 iterations to me. OTOH, it could be 400 iterations, so yes, mea culpa. I am ill-informed. My New Year's Resolution is to stop trying to talk sense to brick walls. Oh! don't do that. Frank's learned where to get a hood for his tri-elmar, and the rest of us have had a bit of fun! It all reminds me of the game of Chinese whispers. Link to post Share on other sites More sharing options...
Theodor Heinrichsohn Posted December 14, 2009 Share #95 Posted December 14, 2009 Funny isn't it.Teddy makes a post which is entirely relevant and presumably factual (unless he is actually a liar - which seems unlikely). . . . . and everyone completely ignores it and carries on ranting. Thank you Teddy for this useful information. Jono, Thanks for your kind comment. You correctly presume that I was being truthful. I did not list the name of the Leica representative, the date and the place, as I was/am not sure that my comment at that time would be appreciated by Leica. Since there were quite a few customers waiting to see the M9 and handle it, I did not have enough time for a more detailed inspection and trial. Teddy Link to post Share on other sites More sharing options...
UliWer Posted December 14, 2009 Share #96 Posted December 14, 2009 ... The great thing about this thread is that I have found somewhere to go and get a Trielmar lens hood and for that I appreciate being involved with the forum. Could somebody please start a thread about Tri-Elmar-lens-hoods? Perhaps we might suddenly find the new firmware update there. Link to post Share on other sites More sharing options...
adan Posted December 15, 2009 Share #97 Posted December 15, 2009 Well, here's an anonymously sourced thread which says "originally planned for Dec. 1 - now put back to 'before Jan. 1'." Which sounds like about what I'd expected. http://www.l-camera-forum.com/leica-forum/leica-m9-forum/109108-new-firmware-possible-improvements.html Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.