Board index » cppbuilder » Re: BorCon C++ Anticipation... making me wait.

Re: BorCon C++ Anticipation... making me wait.


2004-09-20 03:26:41 AM
cppbuilder99
Hi Ron,
Which one? The VCL for VC++ stuff, or me bashing the Borland
management ;-) . The VCL stuff is at www.mitov.com . You can also see
some images of "beat the C++ team" session, but I am not seen, as I was
handling the camera ;-) . You can however take the Harold account for
that ;-)
With best regards,
Boian
BTW: The VCL for Visual C++ is demoed as a VideoLab beta. I have
moved forward since, so if you are interested after the review, you can
contact me for details. Also if you are using C++ Builder with VideoLab
you should install the Visual C++ Beta stuff on a separated system, as
in the beta they share files.
Ron Sawyer wrote:
Quote
Would love to see that - can you point me to a web site?
 
 

Re:Re: BorCon C++ Anticipation... making me wait.

Boian Mitov < XXXX@XXXXX.COM >writes:
Quote
I must have missed you in the crowd.
Rest assured that we did not miss you in the crowd :-)
 

Re:Re: BorCon C++ Anticipation... making me wait.

;-) ;-) ;-) The couple of wine drinks did provide the needed
additional emotional support ;-) ;-) ;-)
Thomas Maeder [TeamB] wrote:
Quote
Rest assured that we did not miss you in the crowd :-)
 

{smallsort}

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
BTW: Where ware you guys at the time? You let me give them all the heat,
and not much backing from you.
You were on such a role, it didn't seem prudent to interrupt you. :)
Borcon was the public's chance to vent. TeamB is usually quiet at these
borcon meet the team sessions, both because it its the public's chance
to talk, and because we have other opportunities to make our views heard.
The TeamB/Borland meetings were immediately after borcon. I saved my
comments for that venue.
H^2
 

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
I'd like to see some major work put into the de{*word*81}. The BCB6 de{*word*81}
still doesn't do things I could do with CodeView or TurboDe{*word*81} a
decade and a half ago. And it's bugs drive me crazy.
The diamondback de{*word*81} contains some nice improvements. The local
variables inspector now updates as you navigate the call stack. They had
one other really nice improvement, but I don't remember it now.
Good news if the BCB project moves forward, totally irrelevant otherwise.
H^2
 

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
An ability to write my own "data viewers" that are views of a complex
data structure.
This is an inherent feature of the .net 2.0 CLR. Not relevant to you, I
know.
H^2
Quote
I'd like to have better control over what happens in multi-threadded
environments. Hitting a breakpoint in one thread will break that
thread, but does it stop other threads?
Currently, it halts the entire process, including threads oustide the
the one that hit the breakpoint. As you step through the code, other
threads are given a chance to run at unpredictable and sporadic times.
The user experience is confusing.
H^2
 

Re:Re: BorCon C++ Anticipation... making me wait.

I am sure you picked it up from where I left ;-)
Harold Howe [TeamB] wrote:
Quote
The TeamB/Borland meetings were immediately after borcon. I saved my
comments for that venue.
 

Re:Re: BorCon C++ Anticipation... making me wait.

"Harold Howe [TeamB]" < XXXX@XXXXX.COM >writes:
Quote
>I'd like to have better control over what happens in multi-threadded
>environments. Hitting a breakpoint in one thread will break that
>thread, but does it stop other threads?

Currently, it halts the entire process, including threads oustide the
the one that hit the breakpoint. As you step through the code, other
threads are given a chance to run at unpredictable and sporadic
times. The user experience is confusing.
I was under the impression that the other threads kept on running when
one was stopped in the de{*word*81}. Hmmm. You're probably right though,
as it's literally been years since I've tested this.
--
Chris (TeamB);
 

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
>The TeamB/Borland meetings were immediately after borcon. I saved my
>comments for that venue.

I am sure you picked it up from where I left ;-)
I tried, but I'm not very good with Russion/Eastern European accents,
especially when swearing. :)
H^2
 

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
I was under the impression that the other threads kept on running when
one was stopped in the de{*word*81}. Hmmm. You're probably right though,
as it's literally been years since I've tested this.
The virtual IP register of debugged process on a single processor
machine doesn't change when you are stopped in the de{*word*81}, so it is
physically impossible for the other threads to run.
I have not tried this on a multi-processor machine.
H^2
 

Re:Re: BorCon C++ Anticipation... making me wait.

You forgot the Chicago, Kentucky, and South Caroline flavors in it
;-) . 'Couse I no have no 'ducation, no nothin', but I gotch your deal
ridh here dudee, and youall be gettin' it ;-)
Harold Howe [TeamB] wrote:
Quote
I tried, but I'm not very good with Russion/Eastern European accents,
especially when swearing. :)
 

Re:Re: BorCon C++ Anticipation... making me wait.

Its 'Carolina' there. Obviously not 'ducated ;-)
Boian Mitov wrote:
Quote
You forgot the Chicago, Kentucky, and South Caroline flavors in it
 

Re:Re: BorCon C++ Anticipation... making me wait.

Tamas Demjen wrote:
Quote
First, the de{*word*81} should work. Currently it only works with a few
basic types (int, double), and even that doesn't always works.
Huh? Every version of the de{*word*81} I've used has supported most types
including user defined classes and enums. The only things it seems to
get confused about sometimes are templates of templates.
--
Andrue Cope [TeamB]
[Bicester, Uk]
info.borland.com/newsgroups/guide.html
 

Re:Re: BorCon C++ Anticipation... making me wait.

Harold Howe [TeamB] wrote:
Quote
The virtual IP register of debugged process on a single processor
machine doesn't change when you are stopped in the de{*word*81}, so it is
physically impossible for the other threads to run.

I have not tried this on a multi-processor machine.
I have no idea how all this is implemented, but all threads are halted
on our dual-CPU dev. boxes as well.
AlisdairM(TeamB)
 

Re:Re: BorCon C++ Anticipation... making me wait.

Quote
Huh? Every version of the de{*word*81} I've used has supported most types
including user defined classes and enums. The only things it seems to
get confused about sometimes are templates of templates.
I use STL too much, as well as shared_ptr, which is a great way of
making the application more stable and preventing terrible bugs, without
using a garbage collector. I'm probably not the average BCB programmer.
However, with large projects, sometimes simple struct members can not be
watched, and that's what caught my attention. It was a plain struct
created on the stack. Anyway, I learned how to live without a good
de{*word*81}, but sometimes I wish it worked better.
I complain more about the project manager, which seems to be
exponentially complex, or heavily polinomial. By adding about 10-20 more
files, the compilation time seems to double (even for very short modules
that normally compile instantaneously with small projects), as well as
the time it takes to add a new unit to the project (can easily take 15
seconds or more to add a single file). This is my absolute show-stopper.
Tom
www.tweakbits.com/tfm/