Jump to content

Qeustion about firmware release number


sdai

Recommended Posts

Advertisement (gone after registration)

Does anybody know why the new firmware shows up as ver.1.91 when it's actually ver 1.092 (as we're being told)?

 

I suspect that 1.92 is Leica's internal release control number but then, does it mean there'll be a ver 2.0 which might introduce some significant changes in various aspects?

Link to post
Share on other sites

I expect there was never any intention to issue firmware other than 1.09, 1.10 and so on but with 1.10 late and the peasants revolting, they thought they needed to do something. They looked for a number between 9 and 10, found there was none and so came up with 91 and 92. The firmware version appears in various places - menu, install prompt, file name - and it looks like the numbering is out of sync in the different places.

 

I doubt we'll get a V2, Leica will be wanting to put the M8 to bed ASAP - there's the M9 and R10 (if it actually exists) to think about and there's no commercial incentive for them to continue messing with the M8 beyond what is necessary to stabilise the product and placate the Great Unwashed.

Link to post
Share on other sites

Guest guy_mancuso

1.09 to 1.10 is a big gap when your updating firmware in pieces which is what they are doing we may see a.93 ,.94 before we see the final 1.10. Which is exactly what i would like them to do. I rather they fix a bug than release it than work on the next project. i am sure they have a nice list to work off of and there going right down the list and fixing as they go to 1.10. Also diifferent engineers are working on different things , one may attack the noise , one may attack the color and one maybe working on the cyan drift. Which I think is what they are doing and I agree this is the way to attack this it will go faster to break it up and break out into teams to take on different solutions. So don't be surprised if there is a 1.93 in two weeks.

Link to post
Share on other sites

They looked for a number between 9 and 10, found there was none and so came up with 91 and 92. .

 

That's very funny, are you sure that they've hired a GERMAN contractor to work on this? thanks for the nice chuckle. :D

Link to post
Share on other sites

Guy, although it might at first seem intuitively obvious to break up firmware development into many small releases, that's actually very inefficient from a broader software engineering perspective. For each release they have to (or at least should!) regression test the entire suite of firmware functions - typically not a trivial task. There are also efficiencies to be gained when a particular piece of code is opened up - it's much better to make multiple changes at one time than to go in on separate occasions, each then requiring execution of the same test suite. For those reasons it makes much more sense to group bug fixes and feature changes into a reasonable basket of tweaks.

 

All the best,

 

Jeff

Link to post
Share on other sites

Guest guy_mancuso

Advertisement (gone after registration)

Yes that maybe better not sure how they are exactly handling that be it each team puts there bit in than run it as one or doing one at a time. But i do think we will see some more updates before 1.1 comes out which i think is good at least they get the feedback.

Link to post
Share on other sites

Notably, just for comparison, many on this forum will be familiar with Ed Hamrick's VueScan. Thought by most to be excellent. Ed periodically does very frequent updates on his software which is greatly appreciated, again by most. If you complain to Ed that you need some 'fix' for your scanner/software combo, it is usually done within 24hrs.

 

In light of a comment above about accumulating all bug fixes together being more efficient, which doubtless is true, the other way makes for very good PR with the customer. Maybe Leica are 'fixing' their PR department with these incremental updates.;)

Link to post
Share on other sites

There is nothing wrong with the numbering scheme you see here. In fact you may see 1.0921 - 1.0922 . . . then 1.093 etc.,etc.,etc. coming our way. The string can be quite extensive from a numerical point of view.

 

As to Jeff's thesis about what constitutes proper testing; that is how some of us do it. Particularly in large commercial environments. But you would be surprised to see how much, sometimes quite critical, software development bypasses (or does not even know about) these time honored practices. I never forget my horror a few years back when I saw whitefrocked technicians change code on a LIVE communications controller at a major telecommunications company.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...