getting bug eyed

Hi folks,
I have been contemplating the state of the SW universe and pondering whether
or not to resume subs for SW2007.
What drove me away in the first place is the bugs and the fact it takes SW
so long to fix them despite the good work identifying them by beta testers
....not to mention the ugly 05sp1 issue...
I find there were 600 bugs in 06sp4 and another 400 are due to be squished
in sp5.
This is not encouraging.
In fact that count late into the release seems to be worse than ever.
Really I would like to see those figures cut to about 1/4 to be content that
SW is on top of the game.
As a solo I really don't want to spend valuable time 'buggering' around with
a multitude of niggling problems and waiting 6-9 months for a solution.
Past experience has taught me not to be an early adopter and to sit and wait
for sp5.
I am thinking of skipping yet another release because of the new splines and
C2 stuff....which I will want to use extensively but just need to work
without hassle and without secretive code changes part way through.

So, how do people rate the current quality of SW say out of ten?
What are your thoughts about how 07 might pan out?

- and please no mindless fan boy responses ;o)
neil

 

Nothing New: I am going to take the low road and wait for SP4-5 before
upgrading my installation. If I pay the subscription fee, that means
they get my subscription money 6 months before I can upgrade, and I
either accept that or let my subscription lapse and decide much later
to reinstate or not.

For a person or company doing intensive work, it has to be a more
difficult decision on when to upgrade, as they have to look out for
problems with their workflow that could cost them more dollars to fudge
around solutions, than the benefits of the upgrade. "You pays your
money and takes your chances."

Given my needs I suspect I will pay the subscription fee, but I will
listen to the early adopters before I decide to write that $2k check or
not.

Bo

 

It's certainly a "damned if you do, damned if you don't" scenario. I
personally chose NOT to send them my money almost 2 years ago, and I
can't say I've regretted the decision although it has led to
inconvenience from time to time. I'm still using 2005 and the bugginess
is relatively low for what I do, which is mostly machine design --
although I've done a pretty fair amount of quite complex geometry using
surfacing for product design (injection molding). I might contemplate
re-upping my subscription sometime in the relatively near future, but
only if my VAR can get me 2006 CDs (even if 2007 has come out by the
time of my re-subscribing). If I can only get 2007 CDs then I'll wait
even longer until 2007 is relatively bug-free ... if it ever gets there.

Regardless, I maintain that SolidWorks Corp is NOT giving sufficient
value for the cost of subscription, and hasn't been giving sufficient
value since before the release of SW2000. Especially with their
re-packaging of the product (eliminating SolidWorks Office simple in
favor of SolidWorks Office Professional) I absolutely RESENT paying them
anything at all.

'Sporky'

 

Neil,
Here is an interesting article from Scienticic American "Dependable Software
by Design".
http://makeashorterlink.com/?N2CB4316D
Software design checking with a tool called ALLOY
Page 2 kind of caught my eye with phrases shown quoted below, then funnily
enough Solidworks crossed my mind.

"What is worse, bugs "fixed" during the testing process often exacerbate
design problems. As programmers debug the code and insert new features, the
software invariably grows barnacles of complexity, creating more
opportunities for errors and inefficient operation........."
"Similarly, bad software tends to get more and more complicated and less and
less reliable, however much time and money are poured into improving it. It
is well known that serious problems with software systems rarely arise from
programming errors; almost all grave difficulties can be traced back to
conceptual mistakes made before programming even started...."

I am thinking of going the other way and bailing out of Subscription.
2006 in my view is a bit of a non event and so far I am 'underwhelmed' by
2007

 

There were three reasons I jumped from Anvil Surfacing to SW in 1996:

1. It was fast
2. File sizes were small
3. It was robust. Very solid and few bugs.

Ten years later.

1. I couldn't possibily say it is fast running on the 166 Pentium I had
ten years ago with 300 Mb RAM. I can't even say it is fast on the AMD64
FX53 I have today.
2. File sizes are huge
3. I spent all day Thursday with a single drawing. Seven crashes later
and whole day gone and the drawing is finally useable. Besides the
interuptions from my own "bug" problems I have to field the bugs from
the users around me. SW is as bad as Anvil was in 1996. And whereas
with Anvil I could usually speak to a developer about problems with SW
I can only speak to a person at the VAR who takes the problems and
starts it down a path with may ot may not preserve the true nature of
the problem as it passes down an endless path to "resolution".

 

About sums up the life cycle of CAD software. I can remember SW97 -

now we're 9 x $1295 dollars later - how far have we got? A better
interface and better configuration manipulation and a few teasers like
animator, cosmorexpress etc but any quantum leap? BOM is still
bolloxs. surfacing is between two stools.

Is there an alternative? Ed hinted at a new CAD software coming out
in 18months. I am all ears for alternative quantum leaps forward.


Jonathan

 

Ourselves waited 'till SP4.1 to jump on SW2006, and got burnt. Lots of
things randomly not working as they should. I may start a list in a while
when I will be sure all that come from the conversion from '04 are gone.
Still, here a lot of people are angry, especialy the least patient of us.

BTW, any idea when SP5 is due? our last hope of getting something usable.
The general feeling is that we should have sticked to '04.

 

We too are in this boat. This is the last time I let the "peanut gallery"
talk me into upgrading.

Without a doubt, it's painful!

 

Thanks for your replies.
Upon reflection I think I will sit out another round.
I am not happy SW have made up any ground and delivered a tool ready for
everyday professional use.
I am a little surprised that relatively few group visitors have taken the
opportunity to voice general dissatisfaction though....seems like a lot of
seat warmers out there )

Aside from the pervasive bugs the only thing that really gets me down me is
dwg performance.
This has to be one of those basic architecture failings Nev.
Somehow SW was not built with real world dwgs in mind - particularly multi
page ones - it is dreadfully frustrating to wait minutes to load and
add/change annotations etc.
I remember someone posting that they actually export the basic dwg and
finish in AutoCAD.
If anyone has some ideas how to get decent speed out of dwgs I would really
like to hear about them....

 

My company has been using SW2006. We were happy with 1.0, 2.0 sucked,
3.0 was worse, 3.1 was ok, 4.0 was buggiest, and 4.1 has been resonably
solid. We pay the subscription and our VAR has been very helpful.

I personally have been using SW since 2002, and I think that my
favorite was 2003. It seemed to be the least buggy, and I was using it
in a large user environment and also doing adminsitration for it, as I
worked for the college IT dept.

Those are just my opinions, but I think that the change from 2003 to
2006 was worth the problems for the added features. I am not sure when
we will upgrade, but probabyl SP1 or maybe 2 at this point.

Personally, I have noticed the

 

... Could the added price for these be justified??

If you haven't check prices lately you might want to.

Software vendors bank on that. The investment in "figuring it out" will be
higher than the the purchase price of the software unless you're buying multiple
seats. Add to that the very significant cost of migrating. That's why there's
so much of a Cola Wars aspect to CAD software marketing.

Maybe there are some local user groups, college classes, etc. where you can sit
and watch the goings on to get a feel for it without investing a lot of time and
effort?

 

That "time" is the four letter word that mucks up the productivity of a
one man shop. I know exactly what John is saying, and it applies to
me, but I only work on my own models, so I have more flexibility as to
whether and when to upgrade.

Bo

 

I must say that I have always been more than frustrated with the
numerous bugs that come with each release. Somehow we have always been
able to work around them or just get by. Right now my biggest
frustration is that I hear the 80/20 license agreement may come to an
end with the release of 2007.
Once upon a time we were just switching to 3D and getting along nicely
with MDT. A coworker comes back from a trade show gushing about SWX. We
knew that Inventor was coming out so we compared Inventor (5 at the
time I believe) with SWX 2001. Without knowing that we would be
receiving a copy of Inventor as part of our MDT VIP subscription, we
chose Solidworks based on its flexible licensing agreement. Due to the
size of our business, we rarely used MDT at the same time. When we did
overlap it was a huge pain to dump out while the other did what they
needed to do. Autodesk absolutely insisted that it be one user at a
time, period! Enter Solidworks. If you both aren't using the software
at the same time more than 20% of the time, one seat is all you need.
Switching from Autodesk to Solidworks was our little way of sticking it
to the man. When we hired another engineer, we did the right thing and
purchased another seat. We still come nowhere close to overlapping more
than 20% of the time. If what I am hearing is true, we may need to get
another seat or switch to network licensing. The sad part is that
Alibre would do everything we need and more, but who can toss out five
years worth of work and change directions now? Nothing like a little
strain on the short and curlies!.

Brad

 

80/20 rule? That's interesting. Sounds like an informal arrangement your
reseller dreamed up. I don't see anything in the license agreement about
an 80/20 rule, it just says that a user who uses the software 80% of the
time it is in use can also install it at home. There is no 20% overlap
clause as I read it.

 

Nice

 

Well I could somehow work around it or just get by...but I refuse!
2000+ bugs per release is not software of a standard for professionals IMHO
)

 

Well I finished my drawing 72 views and 8 sheets later.The actual view
count was up to 105 which shows how may views had to be recreated for
various reasons. At least two days were lost to crash recovery.
Strangely enough saving typically brought on the crashes. I'll have to
admit that for a 2D person to create all the section and detail views
and get them right would have been a challenge. But SW should be able
to do better than this.

The areas where SW really fell down where:

Sections of sections where the first section's alignment was broken
with the parent.
Sections going bad for no apparent reason.
Hatching consistency
Updating cropped views
Random lines appearing on the screen but not on printouts.
The drawing making changes or hanging onto out of date data from the
underlying assembly
Having trouble updating configuration information even thought the
underlying assembly was already updated.
Inconsistent performance. Sometimes acceptably fast and sometimes not.
Inconsistent behavior of extension line attachments
Items sometimes moving on there own (like aligned detail views).
Increased instability after doing a replace operation on a key
subassembly.
Not displaying the detail view letters in the feature tree.

And one or two part related problems like:
Not completely rebuilding and assembly to where the assembly cut
feature had to be opened and closed when SW forgot about it in the
drawing.
Forgeting which surface to extrude up to in a key part.

That's all that come to mind now. There was no point in turning these
in. It would have taken another week just to document them.

As much as I like putting together assemblies and making parts,
drawings still get me down. And in the end I think the drawing turned
out OK. We'll see.

 

Can you give me any insight as to how SW works internally re dwgs?
Somehow to me the whole thing is just very inefficient and consequently
frustrating to use.
It seems like the whole of the SW data has to be accessed, checked and
reworked each time even just to do quite simple things.
I can't see why this needs to be so deep.
A lot of what I want to do could be just a 2d overlay whose position and
scale is linked to the model views position and scale.
Couldn't I have the option to disconnect from the 3d data or have it
suspended for a major rebuild later? - a bit like the old red light
utility...
I really just want to get in there set up some views on some sheets, add a
whole bunch of notes or whatever and get out. This should be able to be done
really quickly but in practice as you note it takes ages and there are
always some things going pear shaped (

 

The Maxwell Render add-in causes a lot of instablity when saving and opening
files. Not sure if you have a copy of Maxwell or not, but it does pay to
disable it when not in use.

John Layne
www.solidengineering.co.nz

 

Try detached drawings then.