Crash, crash, hang, crash, hang,...

Bo, that may all be true but for sure those planes will crash in the ocean from some kind of software bug - right?

A long time ago there was a great article in scientific american about a mathematical theorem that showed it's basically impossible to eliminate every single bug from a software program.

Remember a while back the us navy was testing one of their new destroyers which ended up dead in the water because the winnt based control system software had a divide by zero error. It literally had to be towed back to port!

I don't 'really' want or expect feature developement to stop, my point is they need pay more attention as well to bug fixes. I found a bug yesterday that I 'think' has to do with limit mates and multi-configs that actually corrupted my assembly and drawing file causing me to 'LOSE' work for real (several hours and that was minimized because I had a pack&go backup handy). I havn't had a bug that corrupted a file in solidworks ever (well there was a drawing file in around 2002 that went bad).

I think that developers can't possibly understand how much impact a bug can have on my day/week/month because they are not users. 2006 had a buggy arrow on notes and ballons that probably added 1 million extra mouse clicks to my work during the time I used 2006 - it simply never got fixed. It was an spr from sp0.0. I suppose it seemed minor, but not to me.

Also, I'm not bashing sw - I don't see any sense in that - it's illogical and unproductive. I just want less bugs in this already great program!

Zander

Bo wrote:

Reply to
Zander
Loading thread data ...

I have to agree on fixing bugs instead of constant upgrading which requires more and more hardware.

I discovered a serious problem with 2007 SP1.0, so serious that SWX have esclated it to high priority for SP2.0, why they didn't spot the problem prior to release I don't know. See my previous messge "for anyone using 2007? "

It cost me a approx a week including a full day it took to get may VAR to take it seriously.

Also because the error only surfaced once I started working in 2007 , I now have to re do my work in 2006.

Mo.

Zander wrote:

Reply to
mo

Just wanted to reiterate a point made in this group many times before... Wether you are on support or not you are eligible to receive all of the service packs for the version that you are legitimately using. In other words, if you buy '06 without support, you can still request the '06 service packs since they presumably fix (har) issues with the version you paid for, just not anything past '06. This is something that is not advertised and must be requested of your var but it's something all users should know.

- Eddy

Reply to
Eddy Hicks

Yup, I truly agree on the fact that bug fixes will never end. I did beta testing for a few years and it was painful. I couldn't do it after that, as testing came close to eliminating any productive work.

It is a business, and as such, there is a constant give and take between the "Bug Fix Budget" and the "Income vs Expense" amounts for the business, and the fight is over how much gets put into coding, testing and debugging.

I really do think that the CAD software vendors need to find better ways to set up complex programming systems to eliminate & trap common errors, and minimize others. I don't know who is going to do it, but with the processors getting faster, and software getting far more complex, it seems like some new methods are needed.

I am not a programmer, but just from a problem solving point of view, I would want:

  1. Fault tracking utility built into Solidworks: Users could leave it on or turn it off, as it would likely cause a performance hit.

  1. Key and command logging: Likewise users could turn it on and off.

  2. Emailed fault reporting which allows user input which uses the data from 1 & 2.

  1. Hiring key real-world heavy users to work real life testing on Weekends at commercial pay rates to have new releases run on real world files, that will be segregated from the user's daily work, on a different computer as supplied by SolidWorks. This would be expensive, and for all I know it is already done.

  2. Advanced Programming methods which minimize chances for programming faults: I have read of forms of work & systems for this being done for critical software systems (nuclear...aircraft, etc.).

If SolidWorks, who has hundreds of thousands of active users, looks at the lost time of end users when a bug causes them problems, looks at the impact to users in terms of just that lost time, then it is easy to see that end-user losses can be huge.

If SolidWorks can implement better systems, they may take a performance hit in their profit for a year or two, but what benefit will they then get from their users, which will then result in more users keeping up their maintenance? I have no idea what percentage of SWks users do yearly maintenance, but it would probably go up with more bug free code.

Given the needs to now consider switching the code to run on VISTA, I myself can imagine a whole new set of problems arising for those pioneers who choose to jump into VISTA in the first quarter of 2007. You can imagine that VISTA is loaded with hundreds of features which are not going to benefit a heavy SolidWorks user in his CAD work, and which ought to be offloaded to a 2nd PC.

Speaking for Myself: I would prefer to run SolidWorks on a dedicated computer running an OS that is stripped to the bone, and does NOT run anything not required for SolidWorks. This would allow the OS to be as clean as possible, and not have as many chances for glitches which can result in SWks problems. What is a SWks designer's time worth? Why not give end users the best dedicated tool for the job. If bugs cause a designer to lose 10-20-hours of work a month, how many lost dollars and opportunity for his employer does that represent? Mo just noted a single bug taking up a day with his VAR. It could easily be worth more than a new PC, meaning that dedicating a PC to only that task could pay for itself many times over each year.

There are multiple ways to try to get productivity up all across the board to let users be more productive. I want to see them implemented as best as they can. I would venture a guess that some inventive work in software coding systems, automated testing, dedicated end user paid testing, end user logging & monitoring, optimized PC OS work (maybe with Microsoft's help), and keeping a machine dedicated only to SWks might give substantial benefits to end users of SolidWorks.

Bo

mo wrote:

Reply to
Bo

I've been using swx since '98 and in the last 5 years, I don't even load new versions or take a glance at them until about midway through the SP's. I don't open any realwork files until SP5. I never loaded 2005 at all. It's a fairly painless way to go, but if many users did the same, the bugs would just take longer to be weeded out. If I didn't have to use customer files from newer versions, I'd probably still be using sw2004 and waiting for the mirror assy features to be fixed (sw2013sp4??)

Bill

Reply to
bill allemann

none of those great quality improvements will happen until some real competition emerges in this market. If there are any really good programming suppliers in India, I'm surprised they haven't already published a swx clone. If they were to make a clone that was only a couple years behind in features, etc, but rock solid dependable, they would take half of the mcad business within a few years.

bill

Reply to
bill allemann

I have both Solidworks and Catia, both Dassault products. Catia costs megabucks, Solidworks is the trailer park cousin. Catia is stable, Solidworks not. Seems like money gets you what you want.

--

--------------------------------- --- -- - Posted with NewsLeecher v3.7 Final Web @

formatting link
----- ---- -- -

Reply to
Phil Evans

Speaking of money (regardless of format differences), what does Catia run for a single seat for basic mechanical construction semi-equivalent to SWks in round numbers? I haven't heard anyone mention it lately, not that I am going to change.

Still, Catia apparently delivers on a more bug free code base. SolidWorks can do better.

Bo

Phil Evans wrote:

megabucks, Solidworks is the trailer park cousin.

Reply to
Bo

I would estimate that a Solidworks equivalent Catia v5 version would cost about US$20,000, but of course Catia can be expanded by modules to an awe inspiring cost. eg. aerospace sheet metal module is US$70,000 alone.

Catia is no different to any other expandable CAD system when it comes to snagging you on a reasonable (?) basic cost and them reaming you on the expansion modules.

--

--------------------------------- --- -- - Posted with NewsLeecher v3.7 Final Web @

formatting link
----- ---- -- -

Reply to
Phil Evans

It's to be expected and justifyable to a degree. Not only is there a lack of mass market but the 90 / 10 rule is definitely in effect and that's before the over a barrel surcharge.

Are your licenses rent to use? Are you using it for compatibility or for function?

Reply to
natty

That's incorrect. You pay a $500 Reinstallation Fee plus the regular one year subscription fee. The new term will start the 1st of the month following the order date.

It makes no difference how long the subscription was expired or how may upgrades were released in the interim. You don't have to "back pay" for service - like PTC and UG - nor do you have to pay a separate upgrade charge (like Autodesk and Key Creator.) And you never lose the right to upgrade your software (like Autodesk if you fall more than two releases behind.)

Regardless of other opinions regarding SW Subscription Service, their "reinstallation" policies are the best in the industry.

Reply to
jimsym

That is incorrect! I have the invoice to back up what I have said. Are you employed by SW or reseller? Or was this your best guess.

Kman

Reply to
Kman

"Kman" a écrit dans le message de news: ESa0h.13156$ snipped-for-privacy@newsread1.news.pas.earthlink.net...

My experience is the same than jimsym. I even, for a smaller overall cost, got SW06 without subscription.

Reply to
Jean Marc

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.