Board index » delphi » Borland/DevCO has lost its innovative power

Borland/DevCO has lost its innovative power


2006-10-28 06:55:23 PM
delphi30
I am missing the good old American Power to start things fast without
fear and bring them into the right direction, to make innovation,
business, money.
Borland/DevCO has lost its innovative power!!! There is no real progress
or innovation since Delphi 6 (Delphi 7 = Delphi 6 + some .net
preparations/compiler warnings). Turbo Pascal/Delphi exists since 1983.
Free Pascal exists since approx. 1996. It seems to me that Free
Pascal/Lazarus is making much more progress than Delphi.
In the Delphi Hour people ask often for new features and platforms but
the response from DevCo is: fear, conservatism, carefulness, never touch
a running system, etc.
- Missing language feature
- Generics
- Some concepts from C++ (Static Classes, Class operators, etc.)
- New code generator (Delphi produces still 386 code)
- New Floating Point code generation
- Multi core CPU support
- VCL needs an update !!!
- New layer concepts (Graphic items, Views, etc.) (see Qt Trolltech)
- Unicode
- Linux platform layer
- GDI+
- New graphics formats (PNG etc.)
- Community communication
- Faster fixes (some important QC are open since Delphi 6)
- More code from outside into Delphi (Fastcode, FastMM)
 
 

Re:Borland/DevCO has lost its innovative power

Rod writes:
<snip>
You need some more exclamation signs
 

Re:Borland/DevCO has lost its innovative power

Rod writes:
Quote

I am missing the good old American Power to start things fast without
fear and bring them into the right direction, to make innovation,
business, money.

You know, this is getting very old. I have been a vocal critic of Borland,
but this horse is dead; we've beaten it to death on this board (and
elsewhere).
I'm going to wait and see what happens and quit worrying about what
Borland does with Devco. I think we've all agreed that if it continues
down the path its on, its doomed. I will continue to use the compiler I
have until it no longer does the job I want and then if Borland/Devco
has something that will work for me, I will buy it. Otherwise I will move to
some other environment.
Please people, give it a rest.
Mark
 

Re:Borland/DevCO has lost its innovative power

Quote
It seems to me that Free Pascal/Lazarus is making much more progress than
Delphi.
Some people (myself included) might conclude that this is probably because
it has so much catching up to do. Besides, copying an existing product is
hardly innovative is it?
Quote
In the Delphi Hour people ask often for new features and platforms but the
response from DevCo is: fear, conservatism, carefulness, never touch a
running system, etc.
Hmm, I have no idea what you are talking about, I see new features!
--
Pete
Blessed are the geek, for they shall public class GeekEarth : Earth {}
====
Audio compression components, DIB graphics controls, ECO extensions,
FastStrings
www.droopyeyes.com
====
 

Re:Borland/DevCO has lost its innovative power

Quote
Besides, copying an existing product is hardly innovative is it?
Are you talking about the additions to Win32 Delphi that are in .NET?
Rick
 

Re:Borland/DevCO has lost its innovative power

Quote
Are you talking about the additions to Win32 Delphi that are in .NET?
No. I am talking about someone complaining about Borland not being
innovative and then championing a copy-cat product as his example.
--
Pete
Blessed are the geek, for they shall public class GeekEarth : Earth {}
====
Audio compression components, DIB graphics controls, ECO extensions,
FastStrings
www.droopyeyes.com
====
 

Re:Borland/DevCO has lost its innovative power

Rod writes:
... snip
You know, given it is tone, this kind of contribution to the list is
not productive at all. Personally, I have been using Turbo Pascal
from the very beginning. And after a brief change to C++ (Borland of
course), I have crept back to Object Pascal starting with Delphi
version 2. Every time I start up Delphi, I am confident that
somehow I will be able to solve the problem I want to solve. So far, I
have not been seriously disappointed. And when I have been
disappointed I have had the feeling that it is me as a programmer that
have failed, not the tool itself.
In other words, I think that Borland / DevCo have done a good job for
me so far. Of course, there are features that some users may need,
but there will always be a few cases in the fringes that are straining
for more, more.
For me, the tool is better than good enough, and I suspect that is the
case for the vast majority of Delphi users. For me, every new feature
in Delphi is an invitation and an opportunity, a challenge to learn
something new.
Tom
 

Re:Borland/DevCO has lost its innovative power

Rod writes:
Quote
It seems to me that Free
Pascal/Lazarus is making much more progress than Delphi.
Lazarus hasn't even got to version 1.0 yet. So, I don't know how you
can compare in incomplete product to complete product (with many
feature-rich releases in its history) and say it is making more
progress. I guess if you're talking about Win64 support then it appears
that they've made "more progress". However, we don't know what DevCo
has in the works. Since Free Pascal is an open source project it is
very obvious how far they've got. Which, IMHO, isn't all that far:
sure, it works but even their 32-bit support isn't as good as Delphi
(I've seen benchmarks mentioned before where Delphi outperforms Free
Pascal substantially).
BTW... on the innovation side of things Delphi still has stuff like ECO
that Microsoft doesn't have yet and that came after Delphi 6.
Microsoft's .NET stuff broke compatibility with their older "compilers"
completely while Delphi kept compatibility. that is pretty darn impressive.
Cheers,
Kevin.
 

Re:Borland/DevCO has lost its innovative power

One can argue that BDS2006 wasn't a full version upgrade as well,
with its many bugs and incomplete C++ development environment
upon release. I think open source software has a much different
versioning scheme than most closed source solutions, since they aren't
managed and maintained in much the same way (outside of commercial
variants of open source software).
- Nate.
"Kevin B" <XXXX@XXXXX.COM>writes
Quote
Rod writes:
>It seems to me that Free Pascal/Lazarus is making much more progress than
>Delphi.

Lazarus hasn't even got to version 1.0 yet. So, I don't know how you can
compare in incomplete product to complete product (with many feature-rich
releases in its history) and say it is making more progress. I guess if
you're talking about Win64 support then it appears that they've made "more
progress". However, we don't know what DevCo has in the works. Since
Free Pascal is an open source project it is very obvious how far they've
got. Which, IMHO, isn't all that far: sure, it works but even their
32-bit support isn't as good as Delphi (I've seen benchmarks mentioned
before where Delphi outperforms Free Pascal substantially).

BTW... on the innovation side of things Delphi still has stuff like ECO
that Microsoft doesn't have yet and that came after Delphi 6. Microsoft's
.NET stuff broke compatibility with their older "compilers" completely
while Delphi kept compatibility. that is pretty darn impressive.

Cheers,
Kevin.
 

Re:Borland/DevCO has lost its innovative power

Nathaniel L. Walker schrieb:
Quote
One can argue that BDS2006 wasn't a full version upgrade as well,
with its many bugs and incomplete C++ development environment
upon release. I think open source software has a much different
versioning scheme than most closed source solutions, since they aren't
managed and maintained in much the same way (outside of commercial
variants of open source software).

Hm, partily agree. If you have used D2005 BDS2006 is a big improvement,
starts up faster and crashes substancially less.
And they've integrated C++ where some update was missing for years. They
provided the needed stability update of is as promissed within 2 months
(that is before the end of 2005). BDS2006 was mostly useable from start
where D2005 utterly needed the service pack 1 at least...
Greetings
Markus
 

Re:Borland/DevCO has lost its innovative power

Quote
- New code generator (Delphi produces still 386 code)
Is this really so, and if it is, how bad is it?
 

Re:Borland/DevCO has lost its innovative power

Actually the complier produces 486 code. In most cases, the code it
produces (with a few notable exceptions, which have been QC'd) is fairly
good. So far, within the Fastcode project, we have found very few RTL
routines that benefit substantially from using MMX, SSE etc.
--
regards,
John
The Fastcode Project:
www.fastcodeproject.org/
"Ottar Holstad" <XXXX@XXXXX.COM>writes
Quote
>- New code generator (Delphi produces still 386 code)

Is this really so, and if it is, how bad is it?

 

Re:Borland/DevCO has lost its innovative power

Rod,
Borland/DevCO is a business.
Folks talk about all the features Delphi should have and how dead it
is going to be without said features. However, I never see any
market analysis of the business case for adding those features.
Folks just assume that what they want is what everyone wants.
This is completely untrue. It is interesting that Microsoft is leading
completely in the opposite direction of what most Delphi folks
are screaming for on this forum. No Linux, no native 64, no
optimized floating point, no layering, nor new graphics formats,
etc. Now, you can say that Microsoft is wrong, but...
a) Microsoft has done the research,
b) Microsoft has a way of selling things, and
c) Owning the OS and desktop application market it has a way
of making said technologies defacto standards.
It is generally stupid from a marketting stand-point to bet against
Microsoft. .NET isn't a fad. It isn't a going to go away.
Quote
- Missing language feature
- Generics
- Some concepts from C++ (Static Classes, Class operators, etc.)
- New code generator (Delphi produces still 386 code)
- New Floating Point code generation
- Multi core CPU support
Honestly, would any of that make any difference. Entering the .NET
world with 3GHz processors, we are long past caring about what
code runs behind the scenes. None of this is likely to have any effect
on the sales of the product. We are 10 years past the point where
anyone can publish meaningful statistics about performance in 15MB
applications which spend most of their time in UI, disk, or DB activities.
Quote
- VCL needs an update !!!
- New layer concepts (Graphic items, Views, etc.) (see Qt Trolltech)
- Unicode
- Linux platform layer
- GDI+
- New graphics formats (PNG etc.)
Again, other than Unicode, I doubt any of these have series effect on
sales.
Linux platform... yes, that was wildly successful business case. Always
worth putting a significant number of resources into a 0.5% niche market
where most folks don't want Delphi and won't pay for it.
Others one can get from components or are so obscure that no one
would ever care about them.
Honestly, features are driven by sales potential. The common threat from
everyone screaming at Delphi for not adding their feature is that they
will move to something else... This is a bizare threat since there is
nothing
else except Microsoft products which are pure .NET. Thus, most
of the folks are threatening to go exactly where Delphi (and the entire
industry) are going.
I'm sure there is a business case for .NET. I am sure Borland/DevCO
can say, gee... in five years Microsoft will have spent so many hundreds
of millions moving people there. If we are there, we will make X%
of the market worth X million. I don't see a business cases which
says, if we add Linux (again), GDI+, new image formats, etc. that
we'll get any new sales off this.
Thanks,
Brett
 

Re:Borland/DevCO has lost its innovative power

Nathaniel L. Walker writes:
Quote
One can argue that BDS2006 wasn't a full version upgrade as well,
with its many bugs and incomplete C++ development environment
upon release.
Yip, but it didn't take them long to get up to the release status- and
they fully acknowledged that it was a "preview". The attitude with the
BDS2006 was way different from the 8.0 and 2005 releases. DevCo proved
that they do care about quality and giving users updates and hotfixes if
required.
Incidentally, the original point was that Lazarus/Free Pascal was making
more progress than DevCo. Until we see a final 1.0 product from the
Lazarus team that is a mute point. The only difference between the two
"products" is the approach to software development. Open Source is
simply more transparent. So, it isn't really fair to compare what
incomplete product is available from an open source product vs the
complete products delivered by a commercial entity.
Cheers,
Kevin.
 

Re:Borland/DevCO has lost its innovative power

I totally agree, I think the functionality provided by the IDE is awesome,
and I always hated the multiple floating windows in the old IDEs. I was
just saying that Lazarus does (IMO) function equivalent to a 1.0 version
software. There are many Open Source software packages ready for
production that aren't at version 1.0.
- Nate.
"Markus Humm" <XXXX@XXXXX.COM>writes
Quote
Nathaniel L. Walker schrieb:
>One can argue that BDS2006 wasn't a full version upgrade as well,
>with its many bugs and incomplete C++ development environment
>upon release. I think open source software has a much different
>versioning scheme than most closed source solutions, since they aren't
>managed and maintained in much the same way (outside of commercial
>variants of open source software).
>

Hm, partily agree. If you have used D2005 BDS2006 is a big improvement,
starts up faster and crashes substancially less.

And they've integrated C++ where some update was missing for years. They
provided the needed stability update of is as promissed within 2 months
(that is before the end of 2005). BDS2006 was mostly useable from start
where D2005 utterly needed the service pack 1 at least...

Greetings

Markus