Why is it still impossible to create a working autocad???

[Capt. Computer Crasher]

Well, I wouldn't have described it that strongly. Maybe I have not explored enough. What makes 2008 such a big leap from 2007 or it's previous versions?

Data Linking with Excel, Data Extractor now uses any AutoCAD object not just blocks, Annotation Scaling just to name a few. Granted there are still bugs but these will eventaully will be worked out.

[jguest82179]

With the custom plot routines I obviously don't know how they were created etc in order to troubleshoot the problem, but the problem will still exist when you do decide to upgrade to a future release. It may be worth either taking the time to discover what is actually causing the problem and then look at a way of making your set up migrateable, otherwise every couple of years or whenever you upgrade the same problem will arise.

My system consists of a large number of command line scripts, and that's pretty much it. Yet almost every release Autodesk seems to change something (different things each time) in the way AutoCAD handles plotting that means that I have to go through yet another round of fault finding. When wil they learn - "If it ain't broke, don't fix it!!"

Were the pc3 files used in 2007 actually copied forward from a previous edition of AutoCAD? The variations on the plotting should already be taken care of by future-proofing the code when written. Depending on the language used, it should only be a few changes.

I think they were created with 2007, but I can't be 100% certain on that.

As for "future proofing" the code - it can't get much simpler than a command line script. It's that Autodesk keep changing the nuances of the PLOT command that is causing the problem.

[dgorsman]

My system consists of a large number of command line scripts, and that's pretty much it. Yet almost every release Autodesk seems to change something (different things each time) in the way AutoCAD handles plotting that means that I have to go through yet another round of fault finding. When wil they learn - "If it ain't broke, don't fix it!!"

*smirk* Then they have to deal with users yelling "Where's the new features? We need THIS, and THAT, and the OTHER, and they should all take nothing to implement!". Its a fine line to walk.

As for the scripts - sorry, can't be of much help there. I gave up on those a long time ago; not enough error control and too sensitive to things like acad.pgp customization.

[ideadude]

I find it interesting that you said it takes a little getting used to. We have one user in our office that hates when we upgrade and he has to get used to something new. This year, we researched and tested for about 6 months to make sure we could have a flawless upgrade, and it worked, we actually got complimented on how smooth the transition was from 07 to 08. I hope it can be that easy every year.

Yep, busted...we have a few (including my self) that tend to grumble (I grumble the least and just let out a sigh) when there is an upgrade. I was happy with 2001i.
With Acad 2008 there are some differences that take getting used to.
Thinking back, after the trouble with 2003 and Autodesk changing to a subscription system, I have been less optimistic about the way AutoDesk is operating their business. (slowly bleeding the design community $$$$)
We also test and retest before we get into a change.
Our latest was changing servers with our latest Revit upgrade to 2008 build. We upgraded all our cabling, outlets, connectors, cards, and other hardware to Gigabit. Server to some Raid system. Cost a big bundle of $$$$$
There went my bonus for this year -

[ideadude]

Turn off object fidelity is that setvar chastity_belt set to 1?

I have spent ages producing tables in 2007 recently where the upgrade that's sitting on the shelf in IT would have quite literally saved me hours of draughting time.

Having started on Pafec DOGS in 1988 with a machine the size of a filing cabinet and had 10" floppy disks for storage, modern CAD is fantastic. Its not totally foolproof but unless your pushing what is possible then is it worth doing. Lets all go back to R12 yes its stable on a modern machine but no express tools, no tool palettes no wysiwyg plotting no time saving devises at all.

I was happy with 2001i, No subscriptions, lots of tools and functionality. I still have a copy that I bought when I was consulting running at my home office which I use form time to time. I would be dead in the water if it was a subscription program, cost would be too much for the amount to time that i would be using it.
But now using mostly Revit at work on a daily basis.
I started on a VAX, which was real different, 1980's. Then back to hand drafting until when Acad 8 came out and the firm that hired me was using. Things have changed. For the better for the most part.

[jguest82179]

*smirk* Then they have to deal with users yelling "Where's the new features? We need THIS, and THAT, and the OTHER, and they should all take nothing to implement!". Its a fine line to walk.

As for the scripts - sorry, can't be of much help there. I gave up on those a long time ago; not enough error control and too sensitive to things like acad.pgp customization.

I can't help but get the feeling that you're talking down to me with some unfounded air of smug superiority, and I'm not appreciating it.

Scripts are a perfectly legitimate method of customising AutoCAD. Due to their simplicity and efficiency they are also one of the most effective, too. I have been using them for over ten years, and I will continue to do so for as long as they remain available to use. They are by no means the only method of customisation that I use but for standardisation of plotting they are ideal and I fail to see the point in reinventing the wheel!

At what point did I suggest that Autodesk should refrain from implementing new features? Never! Not once did I say this! What I did suggest was that with something like plotting, which they have had pretty well sorted since R14, they should perhaps leave it alone unless they are going to do a major rework and concentrate more on things like the 3D engine. If they had taken all of the resources that were used to 'fiddle' with the plotting subsystem over the last ten years and allocated them to more useful tasks then we may have seen the implementation of the some more useful features suchs a layer state manager that is actually intuitive to use.

Instead of that, we see through history where something as small as the output filename when plotting to file (and this affects me greatly) in on release including the layout name, in the next release it no longer included it, and then, in a further subsequent release it was back again. Tell me, how exactly is that type of behaviour an 'enhancement'?!?

Surely it would have been simpler to include an option (check box in the dialog and Y/N question on the command line) for the user to choose whether or not to include this piece of data in the filename? Clearly it is not too big of a deal to turn it on or off!!



BTW, if you code your scripts using the full command name rather than the alias then it is not sensitive to pgp changes at all.

[dgorsman]

Whoa there, just expressing a humour opinion (darned if they do, darned if they don't). I'm the one who doesn't use scripts and the intent was to convey that I can't make any comments as to whether they will continue to work or not. Seriously - no insult was intended, even remotely. Sorry to get your double helix in a straight.

[dgorsman]

With some of these mystery problems, I can't help but wonder how many of them are from old items such as pgp, lisp/VBA, blocks, menu files, and other customizations being migrated or outright copied forward without checking to see if the new version actually does things differently. Could be pretty easy to happen, considering staff turnover - people leave, new people come in, and after a few years things get forgotten.

[jguest82179]

Whoa there, just expressing a humour opinion (darned if they do, darned if they don't). I'm the one who doesn't use scripts and the intent was to convey that I can't make any comments as to whether they will continue to work or not. Seriously - no insult was intended, even remotely. Sorry to get your double helix in a straight.

Apology accepted.

I understand the need to find the balance/walk the tight rope, etc. I just really feel that Autodesk has had plotting sorted for a number of releases now, and their resources would be much better spent on things that truly add value (Such as those new features and bug fixes.) rather than fiddling with something that already works well.

It's not like there's a shortage of items on the wishlist each time it's compiled. I can't say I've ever seen a wishlist item that asks "Please rework the order of operations in the command line version of the PLOT command? (Because I just don't have enough problems to solve at the moment! )"

I have found, over the years, that scripts are actually one of the safest forms of customisation as far as future proofing goes, because if you code them with the full command names rather than aliases they tend to stay functional over many releases (except as far as plotting is concerned) as it is pretty rare for Autodesk to drop a command altogether and for the most part when they change how a command works they do it by adding options that can be selected during the command whilst maintaining the original functionality and syntax. My guess is that they approach it this way so as not to have too big of an impact on their users existing customisations.

In fact, apart from plotting and it's associated string of issues over the years, I can only think of one other instance where there was a change made to the way a command functioned and that change had any effect on one of my scripts. I think it was the FILLET command in AutoCAD Mechanical 2006, and it was specific to ACADM. I can't remember the exact details of why it fell over, but I think it asked for the radius every time rather than using a stored variable or something along those lines.

As far as legacy customisations go, I'm pretty lucky here as up until about two weeks ago the only (major) customisations on site were those that I had developed &/or adapted from previous applications so I had a good idea of what would carry forward successfully and what might cause problems.

[jguest82179]

With some of these mystery problems, I can't help but wonder how many of them are from old items such as pgp, lisp/VBA, blocks, menu files, and other customizations being migrated or outright copied forward without checking to see if the new version actually does things differently. Could be pretty easy to happen, considering staff turnover - people leave, new people come in, and after a few years things get forgotten.

Which 'mystery problems' were you referring to, BTW?