in V6.0.0 manual routing got really hard

On 12/16/2011 12:55 PM, Mitch Solomon wrote:

somethings changed, when i manually route and click on the destination pad, if i dont click dead center now... im left with tiny tiny airwires

5.11 didnt do this... i have messed with alot of settings, nothing seems to help

 

Check your catch factor and select factor settings, by default they

should be 5% and 2% respectively. Perhaps on a MAC this is different or

the OS interprets the values differently it may be worth playing with

them to get everything working nicely. Try 7% and 3% or maybe 8% and 4%.

hth,

Jorge Garcia

Guest wrote:

 

On 12/16/2011 12:55 PM, Mitch Solomon wrote:

somethings changed, when i manually route and click on the destination pad, if i dont click dead center now... im left with tiny tiny airwires

5.11 didnt do this... i have messed with alot of settings, nothing seems to help

 

 

Check your catch factor and select factor settings, by default they

should be 5% and 2% respectively. Perhaps on a MAC this is different or

the OS interprets the values differently it may be worth playing with

them to get everything working nicely. Try 7% and 3% or maybe 8% and 4%.

 

hth,

Jorge Garcia

 

It's not that - you can zoom in all you want, and virtually hover over the exact centre of the pad, and it won't "ping" to let you know it's connected.

It even happens if you route a wire into another wire - if the other wire is on the grid, it doesn't think you've finished, so you have to double-click to stop it carrying on.

Just tried it in 6.0.2 (beta version) - still broken.

Steps to reproduce:

1. Move component with air wires connected to pads
2. Manually route to one of the component's pad from another one

It's easier to reproduce if the component's pads aren't on the grid.

No problem!

http://www.tribbeck.com/misc/Test5_11_0.zip is the URL for this.

It's not an exact match in terms of the original layout, but it behaves exactly the same when loaded into 6.0.0 (can't manually route after rotating Q1), and works correctly on 5.11.0.

btw, I used to run the support team for my company (which is also a software developer), so I know how useful it is to have these kind of files for developers. I'm now one of the developers (I was beforehand - just my project got cancelled, and they needed someone to run support...)

On 12/28/2011 1:37 PM, Jason Tribbeck wrote:

No problem, Jorge.

 

Do you still want a V5 version? (Now I'm at home, I've got time to prepare one as well).

 

The only other issue I've got is a MacOS X one, which I posted in the "Is there a list of open issues?" thread. That one shouldn't be too hard to see and fix But I'll leave that for you to comment on that thread instead of this one (I know you're reading this one, so it's really just a bit of a prompt!)

 

Hi Jason,

If you don't mind prepare a V5 variant and put everything together in

the test.zip file that way the developers have access to everything and

we can make they're jobs as easy as possible.

I'll go look at the other thread.

Thanks again.

Jorge

No problem, Jorge.

Do you still want a V5 version? (Now I'm at home, I've got time to prepare one as well).

The only other issue I've got is a MacOS X one, which I posted in the "Is there a list of open issues?" thread. That one shouldn't be too hard to see and fix But I'll leave that for you to comment on that thread instead of this one (I know you're reading this one, so it's really just a bit of a prompt!)

On 12/28/2011 3:49 AM, Jason Tribbeck wrote:

Hi,

 

Okay - it also happens in the Windows version as well. Unfortunately, I don't have a v5 format file (I'll need to switch the directories back over, and I really should be working at work ), but I've made it available (since I'm not NNTP, I don't have your email address, so it's on my web site).

 

http://www.tribbeck.com/misc/test.zip

 

This is a zip of my Eagle project directory, and contains the schematic and board files. It's a simple LED driver circuit that I've thrown together.

 

This is reproducable using the following command:

 

ROTATE =MR45 Q1

 

Then, try and manually route the signal from SV1 pin 1 to pin 1 of Q1.

 

I'm pretty sure it's never happened on 5.x.

 

Awesome Jason,

Thanks for preparing this, I'll forward this to our developers and

hopefully they can get on it for the next beta.

Thanks again for helping, and I apologize for the delay in getting to

you with the release of V6 it's been very crazy around here.

Best Regards,

Jorge Garcia

Hi,

Okay - it also happens in the Windows version as well. Unfortunately, I don't have a v5 format file (I'll need to switch the directories back over, and I really should be working at work ), but I've made it available (since I'm not NNTP, I don't have your email address, so it's on my web site).

http://www.tribbeck.com/misc/test.zip

This is a zip of my Eagle project directory, and contains the schematic and board files. It's a simple LED driver circuit that I've thrown together.

This is reproducable using the following command:

ROTATE =MR45 Q1

Then, try and manually route the signal from SV1 pin 1 to pin 1 of Q1.

I'm pretty sure it's never happened on 5.x.

Hi,

Many thanks for the response.

Since I'm back at work today, I'll need to get these this evening.

However, I have downloaded the Windows version, so I'll see if I can reproduce it here.

The reproduction was a little annoying in that some times I was able to reproduce it easily, started to record and found it worked fine. Or, if I loaded in my example from a fresh start, it wouldn't do it for a while.

But it doesn't take longer than 2 minutes of doing and undoing (which might be part of the problem) stuff before it shows its head - so I don't think it's hard to reproduce.

On 12/26/2011 9:12 AM, Jason Tribbeck wrote:

I'm guessing you're reading this from the newsgroups - in which case, it looks like if you post a video on the e14 forum, the link doesn't come through.

 

For those of you using NNTP, then here's the link:

 

https://www.youtube.com/watch?v=rZvQ3c9uDWk

 

Hi Jason,

Our developers are requesting the following, they would like some

"patients" they could use to test and analyze the issue. My gut tells me

that perhaps having the same design in version 5 and in version 6 with

some clear instructions on how to reproduce the issue exactly. It's not

easy to recreate something like this.

In any event, I will forward you're video to the developers and see if

that helps clear things up, but what will get this issue resolved the

quickest is a couple of test files.

You can e-mail them to me directly and I will pass it on to the developers.

Best Regards,

Jorge Garcia

On 12/26/2011 9:12 AM, Jason Tribbeck wrote:

I'm guessing you're reading this from the newsgroups - in which case, it looks like if you post a video on the e14 forum, the link doesn't come through.

 

For those of you using NNTP, then here's the link:

 

https://www.youtube.com/watch?v=rZvQ3c9uDWk

 

Hi Jason,

Our developers are requesting the following, they would like some

"patients" they could use to test and analyze the issue. My gut tells me

that perhaps having the same design in version 5 and in version 6 with

some clear instructions on how to reproduce the issue exactly. It's not

easy to recreate something like this.

In any event, I will forward you're video to the developers and see if

that helps clear things up, but what will get this issue resolved the

quickest is a couple of test files.

You can e-mail them to me directly and I will pass it on to the developers.

Best Regards,

Jorge Garcia

No problem!

http://www.tribbeck.com/misc/Test5_11_0.zip is the URL for this.

It's not an exact match in terms of the original layout, but it behaves exactly the same when loaded into 6.0.0 (can't manually route after rotating Q1), and works correctly on 5.11.0.

btw, I used to run the support team for my company (which is also a software developer), so I know how useful it is to have these kind of files for developers. I'm now one of the developers (I was beforehand - just my project got cancelled, and they needed someone to run support...)

On 12/28/2011 1:37 PM, Jason Tribbeck wrote:

No problem, Jorge.

 

Do you still want a V5 version? (Now I'm at home, I've got time to prepare one as well).

 

The only other issue I've got is a MacOS X one, which I posted in the "Is there a list of open issues?" thread. That one shouldn't be too hard to see and fix But I'll leave that for you to comment on that thread instead of this one (I know you're reading this one, so it's really just a bit of a prompt!)

 

Hi Jason,

If you don't mind prepare a V5 variant and put everything together in

the test.zip file that way the developers have access to everything and

we can make they're jobs as easy as possible.

I'll go look at the other thread.

Thanks again.

Jorge

No problem, Jorge.

Do you still want a V5 version? (Now I'm at home, I've got time to prepare one as well).

The only other issue I've got is a MacOS X one, which I posted in the "Is there a list of open issues?" thread. That one shouldn't be too hard to see and fix But I'll leave that for you to comment on that thread instead of this one (I know you're reading this one, so it's really just a bit of a prompt!)

On 12/28/2011 3:49 AM, Jason Tribbeck wrote:

Hi,

 

Okay - it also happens in the Windows version as well. Unfortunately, I don't have a v5 format file (I'll need to switch the directories back over, and I really should be working at work ), but I've made it available (since I'm not NNTP, I don't have your email address, so it's on my web site).

 

http://www.tribbeck.com/misc/test.zip

 

This is a zip of my Eagle project directory, and contains the schematic and board files. It's a simple LED driver circuit that I've thrown together.

 

This is reproducable using the following command:

 

ROTATE =MR45 Q1

 

Then, try and manually route the signal from SV1 pin 1 to pin 1 of Q1.

 

I'm pretty sure it's never happened on 5.x.

 

Awesome Jason,

Thanks for preparing this, I'll forward this to our developers and

hopefully they can get on it for the next beta.

Thanks again for helping, and I apologize for the delay in getting to

you with the release of V6 it's been very crazy around here.

Best Regards,

Jorge Garcia

Hi,

Okay - it also happens in the Windows version as well. Unfortunately, I don't have a v5 format file (I'll need to switch the directories back over, and I really should be working at work ), but I've made it available (since I'm not NNTP, I don't have your email address, so it's on my web site).

http://www.tribbeck.com/misc/test.zip

This is a zip of my Eagle project directory, and contains the schematic and board files. It's a simple LED driver circuit that I've thrown together.

This is reproducable using the following command:

ROTATE =MR45 Q1

Then, try and manually route the signal from SV1 pin 1 to pin 1 of Q1.

I'm pretty sure it's never happened on 5.x.

Hi,

Many thanks for the response.

Since I'm back at work today, I'll need to get these this evening.

However, I have downloaded the Windows version, so I'll see if I can reproduce it here.

The reproduction was a little annoying in that some times I was able to reproduce it easily, started to record and found it worked fine. Or, if I loaded in my example from a fresh start, it wouldn't do it for a while.

But it doesn't take longer than 2 minutes of doing and undoing (which might be part of the problem) stuff before it shows its head - so I don't think it's hard to reproduce.