Fair enough, thanks Rob, deleted.  But it does beg the question as from some folks elsewhere about proper testing.  How come links were posted apparently on LM forum (and then reposted here) when it was still in beta.  Too much of a rush? Oh well - we live and learn . . .

I am a bit surprised that LM has folks working on this during weekend!! Good support...!!

How come links were posted apparently on LM forum (and then reposted here) when it was still in beta.  Too much of a rush?

 

I don't know, but most likely just a mistake.

 

As far as proper testing, I've heard it all before and it's full of bad assumptions, inaccurate information, and isn't productive. If you want an accurate idea ... open up the P3D Learning center file ... how many people and how long do you think it would take to test every single feature/item listed in that document?  10 years, 20 year, 100 years ... for "proper" testing.

 

Personally (my opinion which is not representative of anyone other than myself) I'd rather see something before I die ... as opposed to waiting, then die from old age, then figure out how to come back from the dead so that I can see a 100% perfect bug free product.

 

Cheers, Rob.

Good to see all the news is sorted out - I did download the patch from the links here, but haven't gotten to install it.  I was quite confused, as I could not find any info about the release on LM's site.
 

Yep, won't install the patch till the official release is out.

As far as proper testing, I've heard it all before and it's full of bad assumptions, inaccurate information, and isn't productive. If you want an accurate idea ... open up the P3D Learning center file ... how many people and how long do you think it would take to test every single feature/item listed in that document? 10 years, 20 year, 100 years ... for "proper" testing.

 

Personally (my opinion which is not representative of anyone other than myself) I'd rather see something before I die ... as opposed to waiting, then die from old age, then figure out how to come back from the dead so that I can see a 100% perfect bug free product.

++++++++++++++1 from a fellow FS product beta tester.

 

Perfectly said Rob!

Yeah, don't put this one in unless you want a space simulator.  I have no ground textures, all I see are stars.   I hope the last patch can take me back, cause I didn't back anything up.  ulp!

ANYONE HAVE THE LAST PATCH??!!  lol

Oh, never mind.  found it.  Also, I flubbed the install of the patch.  It should not all go into the ShadersHLSL folder.  I'm going to take a nap.

OK, don't install this patch if you like ASN.  Or, if you like ASN with the "Error!" message.  Reverting.

How come links were posted apparently on LM forum

The answer is "know body knows"

 

I always use this analogy when describing testing. Go and write a million page essay without a spell checker. Now have it edited and checked (you'll need a lot of checkers) for spelling and grammatical errors. Will it have errors? After testing will it be more accurate? Will it still have errors? Do you have it checked again? It's your choice as to how many passes you'll accept based on the fact that you won't get a pay check until you release it.

 

I accept that you test software a little differently than editing a million page essay, and when testing software, the answer to the above is usually yes, yes, yes and yes.

 

 

For those that have the patch, I recommend you NOT install.

 

Thanks for watching out for us on this.

Hah, I wasn't going to respond, but no need to snipe. But yes as it happens, I do FS beta as well, and that's after my 48 years in professional IT.  Read that "why all the doom and gloom" .  I've been running with P3D since it started, bugs and all.   I don't mind stuff having faults needing fixing.  In this case LM themselves have said it was a mistake and shouldn't have been released yet.  That's all.   As I put in that other post, P3D is a loads better than predecessors - and a lot of other FS stuff.   Would add, as I'm somewhat older it's also unfortunately more important to me getting the goods, even with bugs, than nothing, quite agree. I have less years left to enjoy it.

 

But purely as a matter of interest, in my professional life we did have to test most everything to death - much, much bigger systems than a flight simulator.  "proper" testing you might want to call it.  Not least because much of it could mean a death or more likely, multiple deaths. Defence systems and simulators, medical and health, air traffic control, fly-by wire, you name it.  Wouldn't have been welcomed in that case if we said we didn't have time.  So we tested everything we could, every single line of code and every item of data we could.  Ask me how long it would take to test etc.?  We used algorithms based on number of lines of code, complexity, coding level (machine code assembler, through all the then extant Cobol, Fortran and specialised high-level languages.  Then we tested all the procedures and instructions in the manuals.  Often hundreds of man years is the answer

 

As I said, never mind, 'tis after all just more water under the bridge

 

BTW referring to the other thread, first computer I worked on had valves in it - but no flight sim  :(  And my daddy is (or rather was) bigger than your daddy.  

 

Things could be worse - but hopefully they won't be.

 

Mike

Apologies, wasn't trying to preach to the converted or "have a go" my fault for not adding that to the end of my post. Was just a for information type post and not targeted at you *blush* As you pointed out, testing is an art form. It's like when the whole of the air traffic computer system came down here in the UK last year I think it was. I bet when they found the fault it was one of those slap yourself on the head moments. I'm sure you'll agree that the most difficult aspect of writing a test script is thinking of the obvious.

 

Edit: valves were magic!

Bri.

Several of my aircraft have the dim VC lights namely  milviz 407 and iFly 737. Hope they get the fix out soon it's a pain flying at night with the dome light on the whole time.

 

 

OK, don't install this patch if you like ASN.

 

Correct for 500 ... if you managed to get 12944, do NOT install it.

 

Cheers, Rob. 

Might this be a good place to lock this thread now as when the final 12944 hotfix is released we will want to install it?

Brian, certainly wasn't aimed at you or indeed any one particular - perhaps more a defence of a view that testing is important than some would have it.  IMHO and (ex)-professional view of course.

 

To add to my previous, with that professional hat on, it generally isn't regarded good practice to release a patch that might actually make things worse.  Whatever the rights or wrongs of adequate testing, it is always important in business software to have a formal release procedure.  And formal quality control is paramount.  My fault, for example I didn't wait to see such a formal release for this patch from LM like the first one had.  I was just about to install it - over trusting, without a backup, when warnings started.   This is business software for LM mainstream and a business user could be rightly unhappy about this. Of course a business user would be unlikely to update on the basis a post here from whoever (though they might from Rob!).  For whatever reason then, this slipped through.  But that's my view anyway.

 

In professional  development I haven't seen anybody not paid because testing took longer, though they might not get a raise if they weren't doing their job properly  :mad:   From the point of view "I'd rather see something than wait forever",  that's fine for amateur users maybe, but there has been a lot of hassle from some developers - I name no names -  about using their products with a professional simulator, precisely because of possible litigation and potential safety risks of incorrect training for example.  Don't mention the  . . .  And this is a professional simulator.

 

Anyway, valves, yes (maybe we should start a new thread on this - it's certainly not P3D) -  and 4096 "words" of  39 bit memory if I remember aright , each bit being a ferrite core nearly two inches across.  All night runs to do some relative simple financial calculations.  Then new-fangled stuff like reels of 35mm magnetic coated film for mass storage. punched cards yet to come on a proper (sorry) computer The ATC system was one of the first systems to have the old 2 foot wide disk platters - where my then "computer manager" thought to test the spin speed by placing a paper clip on the disk and then powering up. Even with a gap between the disks of over an inch, the damage and cost was horrendous.  Those were indeed the days.

 

Hmm, also in the UK. A bit late for an old fella, must go to bed    . . .

 

Edit:  As Brian says, maybe should close this thread, and the Doom and gloom. Start a new old farts thread for  ancient computer reminiscences   - and start a new one when it does happen - "The Official Hotfix 2.5.12944.0 for Prepar3D v2.5 Release Thread"

Not least because much of it could mean a death or more likely, multiple deaths. Defence systems and simulators, medical and health, air traffic control, fly-by wire, you name it. 

 

I can't think of a more complex system than P3D ... that depends on as large an array of changing targets; a wide range of systems, a wide range of tastes in what is entertaining, rather than a decree by law or by intensive research as to what is SAFE, a constantly changing platform - OS, GPU drivers, system capabilites, constantly upgraded specifications for competitiveness in an extremely competitive market- the gaming industry, and a yet a "requirement" that nothing becomes obsolete.  ALL of the systems you mention are on stable targets, most have not been upgraded since a decade before FSX came out, because they are relative to P3D simple and do not need to be upgraded to work with anything else.  There is no other animal like P3D out there.  And as far as Failsafe systems go, LM knows that business too; with them if it's going to be failsafe it's going to be failsafe.