Old bugs back again. Thanks alot CNC!

[MIL-TFP-41]

I did early on John. Found some major bugs and they fixed them. Right now we are waist deep busy. I cant afford down time testing software. I test when time allows. I'd be in HOT water if the bosses found out there machines arnt running cause I cant get things to work as they should in a new release when they work in X5

 

This is the problem with the beta program as it is. All the programmers I know are already to busy doing what they are paid to do (that is, produce bulletproof programs) to mess around with testing unproven software. Lets say you got a hot job that needs to be done yesterday (and I am sure everyone here can relate to that) What are you going to use to program? A proven version where at least you know it's shortcomings.....or a untested version that has a chance of blowing up on you or maybe scrap a part? If I was a shop owner there is no way I would allow beta testing to be done on company time

[MIL-TFP-41]

Woot!!! I made the 500th reply to this thread!!!! and 26,000+ views....ya think CNC Software might be noticing?

[Guest CNC Apps Guy 1]

Woot!!! I made the 500th reply to this thread!!!! and 26,000+ views....ya think CNC Software might be noticing?

But 24,000 of those hits are from the viliage idiot and his minions IP spoofing aliases ... but I digress...

 

 

But Wes, isn't that exactly the point of being a Beta tester?

 

Am I wrong?

 

I know I always run it "expecting" to find things, even when I was programming in a shop.

 

Happy when I didn't, sometimes aggravated when I did

I guess that's te difference between some of us. I use it expecting something to go wrong so I'm on the watch for what I do and how I do it. Is it a slower method up front? Maybe, but I have relatively few problems as a result of the way I approach things so I believe in th elong run, it's ultimately faster, but that's my opinion and y'all are certainly entitled to disagree. Since I'm very deliberate in what I do, how I do it and when I do it, it helps tremendously in repeating it. Case in point, last week I found a bug that happens when you copy a drill operation with tip-comp, then change it to a circle mill toolpath, then try to go turn off the tip comp. It kills it immediatey. I repeated what I did the 2nd try and was able to repeat it, then, on the 3rd tried doing it in a different order and sure enough, I had a perfectly good work around. I'm analytical by nature and it helps me tremendously in my line of work. There are no obstacles, only opportunities to learn different ways to solve a problem. When I was a kid, I enjoyed my Rubik's Cube.

 

I programmed about a dozen 5-Axis turn-keys, about 10 or 11 production turn-keys (mostly HMC 3+1 and a few full 4th), a couple dozen part demos, about 60 time-studies (lathe and mill)... and the times studies I do are the parts that our customer's programmers don't want to do (can't say I blame them 99% of the time). There were zero show stoppers along the way, and 95% of that was on X6 Beta software. Did I have to find some work arounds? Yeah. I'd be lying if I said otherwise. X6 was a no go for the first 2 Betas because the 3D-Connexion drivers were not ready on x64 systems yet, but 3D-Connexion got us something usable in Beta 3. I sesntially program the entire part in a time-study though I cheat on some drilling type cycles if I have to drill and ream a part. The ream cycle has nearly the same amount of motion as the drill cycle. The rest of the stuff, I have to program. It's the only way to get a moderately accurate time-study. So for all intents and purposes I'm programming it. When I do a time study I'm just not posting code. All ym posts are MPMaster based or I use CAMplete (for the 5-Axis stuff). Only one mold though and it was not terribly complicated, though it was done on a 5-Axis so I was using multi-surface 5-Axis stuff and it was good.

[#Rekd™]

James I agree that given enough time and patience most problems can be worked out or around.

 

The problem as I see it is most people/ customers want software that performs well and has little or no problems for the money that they are paying. Is Mastercam that??? I have my doubts with some releases. That said a lot of people have or are using Mastercam to make their $$$.

 

I know you are well versed in a lot of programs and applications as you have put a lot of time and effort into learning these but not everybody has the same time to invest, everybody's situation is different.

 

I can only imagine that being a beta tester is a thankless job and requires a lot of time.

[GoetzInd]

I can only imagine that being a beta tester is a thankless job and requires a lot of time.

 

I think CNC should offer discounts to customers that are active in the beta program. Otherwise where's the benefit? It's obvious that there isn't a wide enough range of active beta testers because a lot of glaring issues are not being caught.

 

Mike

[K2csq7]

I think CNC should offer discounts to customers that are active in the beta program. Otherwise where's the benefit? It's obvious that there isn't a wide enough range of active beta testers because a lot of glaring issues are not being caught.

 

Mike

 

+1

 

I've never been a beta tester, but I can assume it used to be the "cool" thing to do.

I think the "coolness" has decreased at a rate proportional to the increase of "bugs".

 

I think they are going to have to do something different if they want "power" users to continually waste their employers time/$ sending in reports & developing workarounds.

 

 

[mayday]

It's obvious that there isn't a wide enough range of active beta testers because a lot of glaring issues are not being caught.

 

 

Believe me, they are getting caught. We cant make CNC fix them

[Newbeeee™]

Believe me, they are getting caught. We cant make CNC fix them

If this really is the case, then someone at CNC needs to jump on the arse kicking machine

[Matt Berube at Ferron Mold]

If this really is the case, then someone at CNC needs to jump on the arse kicking machine

 

 

My thoughts, exactly. That and I wonder if the software might need some serious re-writing to correct whatever shortcomings are leading to the issues where they fix one problem but that fix breaks 2 other things... If that is what will be necessary it's no surprise they'd be trying to avoid such a huge undertaking by plugging along the same as they always have - fixing what they can and ignoring the root of the issue.

[gcode]

As JM mentioned. the 3D Connexion drivers were not working in the first two betas this time and it think

that reduced the amount of testing that got done

I have been using a Space Pilot for nearly a decade and just flat cannot/will not work without one.

The result is I didn't seriously look at X6 beta till beta3 and I suspect many other testers didn't either.

[Matthew Hajicek - Singularity]

As JM mentioned. the 3D Connexion drivers were not working in the first two betas this time and it think

reduced the amount of test that got done

I have been using a Space Pilot for nearly a decade and just flat cannot/will not work without one.

The result is I didn't seriously look at X6 beta till beta3 and I suspect many other testers didn't either.

 

+1 Insightful.

 

I for the same reason wasn't able to start using X6 Beta until Beta 3, so had very little time to try it out and submit bug reports.

[Guest CNC Apps Guy 1]

The 3D Connexion issue was a show stopper for me. That was not CNC's fault, but nevertheless, there's a LOT of guys like me that depend on that particular peripheral device and I absolutely refuse to work without one, they are that important to work flow.

 

I'm still waiting for 3D to get their SDK to CAMplete... In CAMplete no 3D Connexion device is more of a nuisance. SHould have it soon though. x64 really screwed 3D up. The CATIA drivers were even hosed for a bit.

[neurosis]

Believe me, they are getting caught. We cant make CNC fix them

 

 

I would be curious to see how they prioritize bugs. Are some bugs that have been around for several releases considered low priority because they only affect certain people? Are they considered low priority because new functionality is more important?

[Bill Henderson]

I really can't spend the time reading all of the posts on this subject, but to paraphrase the complaints where do most of the "bugs lie"? Is it something seen more in Lathe, Mill, 5 Axis, or design side? Are they critical as in system crash or worse tool crash or are they more on the interface side with functionality?

 

I believe the priority list is sort of set that way.

[Guest CNC Apps Guy 1]

I've never asked how bugs were prioritized... would be cool to know.

[Mic]

I've never asked how bugs were prioritized... would be cool to know.

 

My guess is:

 

Number of customers reporting the same bug

How critical it's

How long time will it take to fix it

 

Not sure about the order

[Bill Henderson]

I read to page 6... got bored. Like a bad movie you really didn't want to watch but was interested in the ending...

 

Was clock speed an issue?

[Bill Henderson]

Am I the only one that rarely has to go back and change tool numbers?

 

Matt, James, or Mr. Paris... do you think I should move onto X6? Thinking about downloading it, but some hesitation? What do you guys think? Not a hijack. Email me or send me a message.

[Thee Rickster ™]

Am I the only one that rarely has to go back and change tool numbers?

 

I have to change tool numbers, in toolpath groups, up to 10X aday

[Aeroguy]

Had something weird happen to me recently. I had to change something in a toolpath, and after I accepted it, a number of totally unrelated toolpaths went dirty. Ok says I, and just recalculated them. Well unknown to me, aside from recaclculating them, they also changed the tool number for those particular toolpaths. So when I posted the whole thing out they had the numbers all wrong, and I totally scrapped the part because it obviously used the wrong tool for those particular paths. When I went back into MC and rooted through them all I found the problem.

Why the heck would MC do this ??

[Bill Henderson]

I have to change tool numbers, in toolpath groups, up to 10X aday

 

I see. I use multiple machine defs for various wcs offsets. not so much toolpath groups. But I can see that being an issue.

[peon]

Bill, for the most part, bugs you may have had in X through X5 may still be in X6 and that was my main bitch. I think it is certainly worth installing X6 because the crunch speed and verify is much faster with the 64-bit suppport. The stock model works well for most of my roughing ops, but can sometimes crash when finishing or display incorrectly with some custom tools. I just avoid using the stock model for most of my finishing ops unless I'm working with a small file. The bullnose support in the dynamic and optirough paths also work rather well. If you use the X+ Setup Sheet, you will have to wait till the new version is released since the X5 utility won't work with X6.(That is a deal breaker for several users I have talked to.) Just get used to working around the bugs like you have in the past and lots of good folks on this forum are here to help.

[neurosis]

Bill, for the most part, bugs you may have had in X through X5 may still be in X6 and that was my main bitch. I think it is certainly worth installing X6 because the crunch speed and verify is much faster with the 64-bit suppport. The stock model works well for most of my roughing ops, but can sometimes crash when finishing or display incorrectly with some custom tools. I just avoid using the stock model for most of my finishing ops unless I'm working with a small file. The bullnose support in the dynamic and optirough paths also work rather well. If you use the X+ Setup Sheet, you will have to wait till the new version is released since the X5 utility won't work with X6.(That is a deal breaker for several users I have talked to.) Just get used to working around the bugs like you have in the past and lots of good folks on this forum are here to help.

 

 

Ive had several crashes while trying to use the stock model. While trying to edit operations and also while trying to run the verifier. As of now, the stock model has barely been usable and non-trustworthy for me.

 

Peon,

 

Youve almost got the record. <grin> probably a couple of more days.

[peon]

Ok, just crashed regenning a previous stock model on a roughing op after making a change to the depth limit. Hahahahaha!!!

[Bill Henderson]

I can't say I would use the stock option for what I do here. When I used to program large molds I could have used that. Most parts I cut now can fit in your hand or in some cases a thimble.

 

Thanks for the replies.

 

I did beta testing back in 2001 or 02? yes it is a thankless job, but you do get to use the latest and greatest along with any bugs you come across.

Maybe when I joined the forum. who knows?