SolidWorks Getting Slower with Age?

I did some tests with my "Ship in a Bottle" benchmark between 2001Plus, 2003 and 2004 to see what the differences were.

I'm posting my results for whatever reason.

Hit CTRL-Q three times before running macro. Times are in seconds NO add-ins are enabled Verification on rebuild "Off" HLR Edges "Off" Fast HLR/HLV (Draft Mode) "On" High Quality for normal and dynamic view "On" No re-boots

50 rebuilds using the included Rebuild Macro WinXP Pro SP1 Who cares what my computer hardware is (it's old)

2001Plus native file (saved as 2004) Test done in 2004 SP2

74.08 1st try 73.96 2nd try 73.86 3rd try

2001Plus native file (saved as 2003) Test done in 2003 SP4

70.32 1st try 70.27 2nd try 70.05 3rd try

2001Plus native file Test done in 2001Plus SP5.1

66.52 1st try 66.37 2nd try 66.13 3rd try

I also wanted to see if a native 2004 file would make any difference...

2004 native "re-created from scratch" file Test done in 2004 SP2 73.78 1st try 73.66 2nd try 73.51 3rd try

If anyone wants the 2004 file, let me know and I'll post it. Note that if I open any version at random and perform the test, the results are the same. In other words, a re-boot is not necessary between tests. I would never re-boot in a real-world situation anyhow.

Mike Wilson

----== Posted via Newsfeed.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= 19 East/West-Coast Specialized Servers - Total Privacy via Encryption =---

Reply to
Mike J. Wilson
Loading thread data ...

Thanks Mike! You again confirmed my results using Spec's benchmark. 5-10% each and every version. Again, for those math-limited, that's about 2-4 hours per week to do the same tasks. And don't even pretend that SW2003 -or- 2004 is saving you 2-4 hrs per week with all the new bells and whistles (read: software demo's).

- Eddy

Reply to
Eddy Hicks

Well, thanks to both you and Mike, it makes me feel better about my test as well.

.. 8^)

Eddy Hicks wrote:

Reply to
Paul Salvador

Mike,

Thanks, this is the kind of justification people need to convince the boss not to upgrade. It is objective, repeatable and not subject to much interpretation. And the test is open source. No hidden gimmicks.

By the way I just used your iterati> I did some tests with my "Ship in a Bottle" benchmark

Reply to
kellnerp

I almost decided to give in and pay the maintenance for this year. I almost figured it would be me who loses in the end since they'll charge me even more later. But with the trash they've been putting out I decided nope. Maybe if some of us stick to our guns and hold back they'll get a message. And if not, it'll be that much easier to switch to something else later this year, since I'll have more to spend :) Once again, let me say that my VAR is top-notch, and I love the software. But I eat crap for no one, no matter who's serving it. For the sake of all of us and our VAR's I hope things start to improve soon.

This is the first time in 8 yrs I held back maintenance fees. The company back then was Autodesk :)

- Eddy

Reply to
Eddy Hicks

Cool!

----== Posted via Newsfeed.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= 19 East/West-Coast Specialized Servers - Total Privacy via Encryption =---

Reply to
Mike J. Wilson

This would be the case if you were always waiting SW but in my case (I am a designer) I spend most of my time thinking rather than modeling. Modeling just come once the idea is more precise and I do it to fix dimensions and details. However for a draftman or somebody that just modelize in a days I understand.

I don't thing I loose 2-4 more hours since 2004 but I really like the new features, they are time saver for me.

Reply to
Robin Boudreault

I have noticed the main area of slowdown is the 2d drawing.

Reply to
Rocko

I agree that drafting and sketching are the most visually noticeable performance problems but I think the file saves, opens, etc. are taking extra time. Likely due to the larger files.

- Eddy

Reply to
Eddy Hicks

Ok, it's taken me awhile but I ran your ship with the parameters you used:

SW2004sp2.0 (it's for testing) Athlon64 3200+, 1Gb ddr400, QuadroFX1000, 120Gb Raid-0 HD

(your downloaded file saved as 2004sp2) = 29.14063 = 29.15625 = 29.14063

- Eddy

Reply to
Eddy Hicks

Nice! We finally broke 30 (and with 2004) woohoo!!

Mike

----== Posted via Newsfeed.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= 19 East/West-Coast Specialized Servers - Total Privacy via Encryption =---

Reply to
Mike J. Wilson

You do realize that it is just a matter of time before SW steals a page from the video card manufacturers' playbook and starts optimizing SW to perform well on the 'ship in the bottle' benchmark....

Jim (mostly kidding) S.

Reply to
Jim Sculley

AHHHHHHHH !!!!

Reply to
Eddy Hicks

Hey, how did you know about that? I saw some articles on HardOCP a while back about benchmark cheating and how the vid card people tweak their drivers to look good in the tests.

So yes, that thought actually did cross my mind!

Mike

----== Posted via Newsfeed.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= 19 East/West-Coast Specialized Servers - Total Privacy via Encryption =---

Reply to
Mike J. Wilson

Damn, so the theory here could be that they tweaked the software to make Spec's benchmark or your benchmark look better for 2004. And yet, it's still 10% slower than 2003 running the same benches. Har :) It's funny to think about that actually being a possibility.

:)

- Eddy

Reply to
Eddy Hicks

I would like to try this....where do I get the files and macro from??..sorry for the stupid questions,lol...

mail to cd@"NOSPAM"cdignition.com

Thanks

J

"> >

Reply to
CDignition

Here you go...

formatting link
You can't miss it. The macro is included in the ZIP file.

Enjoy, Mike Wilson

----== Posted via Newsfeed.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= 19 East/West-Coast Specialized Servers - Total Privacy via Encryption =---

Reply to
Mike J. Wilson

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.