What's the largest Creo feature count you've seen?

[dross]

I tried adding a couple of parts to a layer. Tried another with an extrusion from a part in the layer. Both layers have the same single option "info" when right-clicking in the model area.

 

[tobbo]

sorry, my bad!

You have to highlight the features that is in the layer (not the layer itself. so just expand the layer and click on the first feature, and hold down shift and click on the last so every feature is highlighted) . Then go to sketching area...

The main benefit of this would be that you don´t have to search for single features/pattern in the modeltree , you will find all of them at one place (the layer)

//Tobias

 

[dross]

Since my patterns are easy to find and your method involves more work than the way I'm doing it, I'll continue as is.
Thanks for your input.

 

[CPiotrowski]

I was at a company last month that claimed to have the largest # of features in a Creo part. They were 10k off. This chair from Herman Miller has the largest I've seen. Over 21k # of features. Check it out. One of my past participants in training did this one w/ their India partner.


Sayl - Office Chair - Herman Miller < just wanted to shake it up some today.

Im in Ohio near the Indiana Boarder this week teaching classes BTW

Seems like an enormous amount of features to make that chair. I did a missile silo control box some years back that was about 5,000 features (not counting features within patterns) and it was far more elaborate than that chair. From my experience working with other peoples models...feature count doesn't necessarily express the complexity of the model but rather the lack of knowledge of the system and tools to produce a part with the fewest number of features. Cool chair tho...I'd be interested to see how many features it would actually take to make it.

 

[dgs]

... I did a missile silo control box some years back that was about 5,000 features (not counting features within patterns) and it was far more elaborate than that chair. ...

Why wouldn't you count the features within patterns? They're required to make the part, right?

I care little about minimizing feature count and care more about a logical feature breakup. Turned parts are a good example as people tend to create a single sketch for the entire part or most of it. I'd prefer to revolve the major body and then create separate revolves for each o-ring grove or other functional aspect of the part. I feel it makes for an easier to understand and more flexible model. Same goes for screw bosses in a plastic part. I may be able to extrude all 15 of them in one feature, but if there are 5 parts being attached with 3 screws each I'd rather break them into 5 features.

 

[CPiotrowski]

Why wouldn't you count the features within patterns? They're required to make the part, right?

I care little about minimizing feature count and care more about a logical feature breakup. Turned parts are a good example as people tend to create a single sketch for the entire part or most of it. I'd prefer to revolve the major body and then create separate revolves for each o-ring grove or other functional aspect of the part. I feel it makes for an easier to understand and more flexible model. Same goes for screw bosses in a plastic part. I may be able to extrude all 15 of them in one feature, but if there are 5 parts being attached with 3 screws each I'd rather break them into 5 features.

They are required to make the part but the feature you used was the pattern itself, not all the instances within the pattern. When counting features for raw model complexity...adding instances within actual features is a misrepresentation. If someone tells me they have a 5,000 feature model without patterns...that's a hell of a lot more impressive than a model with 100 actual features and 10,000 when adding the instances within one pattern. I agree with you on people making sketches too busy or complex...the sketches should be clear and separated for the ease of showing dimensions on a print...basically intuitive design intent. From a personal perspective...I get frustrated when I get models from people who don't put consideration into the amount of features they are using because it makes working with that model far more confusing. A lean, streamline model tree should always be the goal.

 

[CPiotrowski]

Let me clarify...
If I were looking at number of features when talking about regeneration time then I would include pattern instances.
If I were looking at number of features when talking about the complexity of a model and time to make it...I would not include pattern instances.

Just depends on what is being discussed and/or compared.

 

[tobbo]

Cpiotrowski, nice to see that your "back", haven´t seen you for a while? Didn´t you model a ford mustang a long time ago and made some pretty nice renderings using pro/E Arx?

//Tobias

 

[CPiotrowski]

Cpiotrowski, nice to see that your "back", haven´t seen you for a while? Didn´t you model a ford mustang a long time ago and made some pretty nice renderings using pro/E Arx?

//Tobias

Good to see you too Tobias...yeah, that was me, good memory...I think that rendering q&a thread is still kicking around here somewhere. That's still one of the longest threads I've seen on a forum. There was a lot of interest back then for native renders in Pro-E. I've definitely come a long way since then. It's been years since I've had a computer at home thus the reason all my work suddenly stopped. I've got one now and plan on re-igniting some modelling and rendering stuff on here.

Here is that Mustang render you mentioned for a trip down memory lane. The model was made more than 10 years ago...man time flies!

 

[design-engine]

maybe we can get some traction back on this forum. For those of you who took that design engine surfacing class you realize we don't care about feature count either. Actually more features can be more simplistic depending upon how you define complicated. For me it's being able to change a part 20 times in 10 minutes.

 

[mgnt8]

Where it makes sense to have a lot of features is the finishing work - draft, rounds, chamfers.
It's easy to pile a bunch of edges into one round feature and then lose track of what was in there if something fails. I've found it's best to split them up and group them.

It's worse for draft because if one surface fails, they all disappear when you go to redefine, so you have to really keep track of what you had in there. The draft feature failing is one of the worst long term bugs that PTC never addressed.

 

[dross]

maybe we can get some traction back on this forum. For those of you who took that design engine surfacing class you realize we don't care about feature count either. Actually more features can be more simplistic depending upon how you define complicated. For me it's being able to change a part 20 times in 10 minutes.

Unless it takes 30 minutes to re-gen...

 

[design-engine]

I remember the strategic coffee breaks back in the early 90's there were 30 min regenerations all the time. Anyone remember the 'clock' and explaining to your manager that the computer is thinking? The Sun spark 2 1992 with 32 megs of ram running SUNOX

 

[mgnt8]

You were lucky if you got a Sparcstation. I remember being saddled with an IBM Power(less) PC. They should have moved my desk next to the coffee machine.
On the days when I'd open the master assembly, I'd just take a tour of the whole tech center.
That might have been when I started dating the VP's receptionist. Eventually, I got upgraded to an SGI Indigo and the relationship fizzled.

 

[per2fh]

mgnt8
I take it the Indigo allowed faster access?!

 

[design-engine]

HAHAHA... I remember doing push-ups at the coffee machine w/ other contractors 1994