Board index » delphi » Re: Already "too late" for x64 development

2005-08-05 11:31:36 PM
delphi185
Eric Grange writes:
enterprise market for the next 2 years. (I'm assuming that enterprises
replace 1/3 of their machines every year.)
--
Mike Swaim XXXX@XXXXX.COM at home | Quote: "Boingie"^4 Y,W & D
MD Anderson Dept. of Biostatistics & Applied Mathematics
XXXX@XXXXX.COM or XXXX@XXXXX.COM at work
ICBM: 29.763N 95.363W|Disclaimer: Yeah, like I speak for MD Anderson.
 

Will DeWitt Jr. writes:
size could change, and didn't want to break code. (And that is why C is
littered with odd types such as time_t.)
I've worked on a fairly large application that could be compiled
either 32 or 64 bits by just switching compiler/libraries from 32 to64
bit.
--
Mike Swaim XXXX@XXXXX.COM at home | Quote: "Boingie"^4 Y,W & D
MD Anderson Dept. of Biostatistics & Applied Mathematics
XXXX@XXXXX.COM or XXXX@XXXXX.COM at work
ICBM: 29.763N 95.363W|Disclaimer: Yeah, like I speak for MD Anderson.
 

On Fri, 5 Aug 2005 11:14:48 -0400, "Dennis Landi" <XXXX@XXXXX.COM>
writes:
to supporting mono, that looks like the best option."
You agree with his analysis but won't be using more boutique tool
sets.
I think that this is known in the industry as a 'circle jerk.'
I'm beginning to have a lot more sympathy for Kaster.
-ckd
 

"ckd" <XXXX@XXXXX.COM>writes
I, personally will be looking closely at Qt 4.0 on VC++ (or whatever the VS
IDE is called these days, whidbey or whatever). At least the leaves the
door open to native cross-platform development.
For normal "I.T. consulting", I guess I will be embracing C# but that won't
have anything do with my own products.
-- d
---------------------------------------------------
Need to see what's happening?
Check out the Delphi Community Blog Aggregator
delphi.flashblogger.com
 

There are so many mistakes in your assumption it isn't even funny.
Not every corporation is a hospital. I know lots of corporations
do rolling roll outs to different departments. So one department
might get all new 64 bit machines one month with all the machines
running new 64 bit software.
With all things being equal, I'd suspect most companies would
deploy the same 32bit software they are using currently on 64bit
machines (64bit version) as soon as the software is available.
Mike Swaim writes:
Thomas Miller
Wash DC Delphi SIG Chairperson
Delphi Client/Server Certified Developer
BSS Accounting & Distribution Software
BSS Enterprise Accounting FrameWork
www.bss-software.com
www.cpcug.org/user/delphi/index.html
https://sourceforge.net/projects/uopl/
sourceforge.net/projects/dbexpressplus
 

I am not using Chrome for anything yet. Everyone seems happy for me
to use C#. On the other hand, there are some real issues with C#
and handling simple data types. For our own internal stuff, I
am looking really hard at Chrome. We haven't decide to drop Delphi,
but it is becoming more of a real world business decision every
day. I can not in good conscious blindly support Delphi any longer.
ckd writes:
Thomas Miller
Wash DC Delphi SIG Chairperson
Delphi Client/Server Certified Developer
BSS Accounting & Distribution Software
BSS Enterprise Accounting FrameWork
www.bss-software.com
www.cpcug.org/user/delphi/index.html
https://sourceforge.net/projects/uopl/
sourceforge.net/projects/dbexpressplus
 

Thomas Miller writes:
of either did a variation of replacing a machine every 3 years, or
replacing it when the machine was effectively dead.
possible to keep life simple. If half your machines can run version a,
and all can run version b, then you will probably want to go with version
b for everybody.
--
Mike Swaim XXXX@XXXXX.COM at home | Quote: "Boingie"^4 Y,W & D
MD Anderson Dept. of Biostatistics & Applied Mathematics
XXXX@XXXXX.COM or XXXX@XXXXX.COM at work
ICBM: 29.763N 95.363W|Disclaimer: Yeah, like I speak for MD Anderson.
 

The argument is always 'Borland is making money and gaining knew
customers'. All I see is that people have bought the last 2
versions to see what is going on technology wise and then put
it on the shelf to collect dust. I am sure there is a small
percentage actually using Delphi 8 and Delphi 2005 development.
I bet there is more not using it for real programming, more then half.
Our user group meetings have gone steadily from 35 people each
month to about 15 on average. Not one person at the user group
is doing any mission critical programming with D2005 in our
user group.
So Borland can continue to put their heads in the sand or take
notice that one of the programmers that has supported them
strongly from Delphi 1 to Delphi 7 is looking else where for
his development tools. I am not the only one doing this.
It is because they have mismanaged the stewardship of the
best development tool on the planet (as of Delphi 7).
Sad.
I am not sure where they get their marketing information, but
they need to get it some place else. It is also evident that
since Delphi 6, the only thing Borland upper management has
been interested in is milking Delphi for revenue while it
steadily declines. That is the only reason you don't give
sales people commission on a product. You aren't looking
for new sales, just repeat business.
I hope it is spun off into its own company. I hope the new
owners will spin off the VCL for open source so it will
progress faster. The IDE and VCL should be on separate
release cycles so the IDE can take its time being released
and the VCL can be fixed quickly and dynamically.
Billb writes:
Thomas Miller
Wash DC Delphi SIG Chairperson
Delphi Client/Server Certified Developer
BSS Accounting & Distribution Software
BSS Enterprise Accounting FrameWork
www.bss-software.com
www.cpcug.org/user/delphi/index.html
https://sourceforge.net/projects/uopl/
sourceforge.net/projects/dbexpressplus
 

"Thomas Miller" <XXXX@XXXXX.COM>writes
worse, I think thier competitors have gotten better.
I also use Delphi 7 for Win32 programming (our primary application) and VS.NET for
.NET programming (all future projects). My comparison was between Delphi 8 and VS
2003. If Delphi 8 had been "clearly superior" for my development (as Delphi has
been for Win32) that would have been one thing. Turns out VS was actually
more stable and so the decision was made.
I see very little _benefit_ to using Delphi 8 (and presumably D2005) over VS for
_my_ applications. I don't think this is so much an indictment against
Borland, but rather a testimony to MS's improvements in VS. that is my take
on it, anyway.
 

Mike Swaim writes:
I would bet over half (more then 50%) will do rolling roll outs.
You hear of this happening with Linux changes in government all the time.
--
Thomas Miller
Wash DC Delphi SIG Chairperson
Delphi Client/Server Certified Developer
BSS Accounting & Distribution Software
BSS Enterprise Accounting FrameWork
www.bss-software.com
www.cpcug.org/user/delphi/index.html
https://sourceforge.net/projects/uopl/
sourceforge.net/projects/dbexpressplus
 

Mike Swaim in <42f39bad$XXXX@XXXXX.COM>inscribed:
this year is 32 bit so I will not even have the chance of a 64 bit machine at work until 2008.
This replacement rate is more rapid then for most of my previous jobs (4) and much more so then
for my current clients.
 

"Thomas Miller" <XXXX@XXXXX.COM>writes
The argument that Borland probably would not find it profitable to release a
64-but compiler right now really has nothing to do with whether they are
currently making a profit or not on the results of their prior decisions,
because it is about whether such an action itself would be profitable,
ceteris paribus. In fact, one could easily argue that the quarterly loss
would have been even worse if Borland had been diverting resources away from
current revenue sources in order to build a 64-bit compiler. It is a luxury
they probably could not have afforded.
Near as I can tell, your argument seems to be that nobody can say that a
64-bit native compiler would be unprofitable as long as Borland itself is
unprofitable, and that is just plain silly.
 

"Thomas Miller" <XXXX@XXXXX.COM>writes
 

"Eric Grange" <XXXX@XXXXX.COM>writes
applicable to the market as a whole, because if it was descriptive of the
market as a whole, then only the one fastest CPU ship or PC would garner all
sales, and these sales would be equal to the entire existing PC stock of
slower systems plus newcomers to the market, every time a new faster system
came out. We all know that the best selling items at vendors of PC's is
almost never their fastest system. So it looks like reality dares to differ
with your theory.
 

Dennis Landi writes:
64-bit compiler, and trusts Borland's research.
Why would you expect him to support the Escrow? Your terseness seems kind
of negative in this instance.
I'd support the Escrow, myself, even if I don't believe in the 64-bit
compiler, for two reasons:
1) It'd give Borland an opportunity to live up to it is promises. They've
said, in one form or another, give us enough money, and we'll build it.
2) A formal escrow project, kind of like the UK's No2ID thing, would allow
all the 64-bit advocates out there, and the lurkers, the opportunity to
"put up or shut up". If your business really would benefit from this
thing, now, then the small investment per person shouldn't be that hard to
part with. Certainly cheaper than moving your codebase to VS.
<tongue in cheek>
Honestly, I think if you do set it up, #2 might persuade Nick & others who
don't believe to promote the escrow, if only to shut you guys up.
<tongue in cheek>
Now I am wondering if that will be taken as putting words in someone else's
mouth, or insulting, or whatever, but it is not meant that way. I've
suggested the escrow account before. The newsgroup wars are not moving
Borland to do anything that we can see. Dangling cash in front of a
public company is pretty significant. Use your antagonists against
themselves.
Just imagine the media attention you could get. You might even get to
ride Coates' coattails.
-Brion