Thursday, March 24, 2011

Delphi Down Under

This year’s Australia Delphi User Group symposium is a two day event with sessions in Melbourne and Sydney. The first of those – the Melbourne session – happened today. We met in the John Scott Meeting House at the La Trobe university.

Turnout was quite big – nearly 80 very attentive listeners.

The sessions started with David I talking about what Embarcadero offers now and what are its future directions; about what Delphi offers at the moment and – most importantly – what will 64-bit Delphi bring to developers and where can we expect problems (mostly from having to use 64-bit Windows API and from pointer size change to 8 bytes).  There will be internal 64-bit assemblers although we will not be able to mix pascal and assembler code in one method. There will not be an extended type anymore :( If that bothers you (it surely bothers me), send a letter to your local Delphi representative.
David I
After short tea break, I spent quite some time speaking about multithreading in general and about high-level multithreading support in OmniThreadLibrary (Join, Future, Pipeline, Fork/Join and Parallel For). The session was well accepted and it was really nice to speak to such attentive and responsive public.
Primož Gabrijelčič
(photo by Andrea Coffey)
After a longer break for lunch, Thorsten Engler presented overview goals and implementation details of the NexusDB database engine and of the memory manager included in the NexusDB. Although at some places very technical and detailed, the talk was very informative.
Thorsten Engler
The last speaker was Lachlan Gemmell with an exhaustive list of language improvements since Delphi 7, nicely tabulated by releases – an excellent resource for all Delphi 7 programmers out there!
Lachlan Gemmell
While the symposium concluded with a prize draw (and there were almost as many prizes as there were attendants), we speakers were flying to Sydney where we will repeat our presentations tomorrow.
prize draw


  1. I personally think it's wonderful news that Extended will map to Double.

  2. And what about all the people that have extendeds stored in data files? Or even worse - in a massive database? And all the people doing simulations? They really need 64-bit to access all the memory AND extended precision.

  3. On the Sydney session, David told the audience that there are supposed to be conversion routines for extended included in the Delphi. That alleviates the pain a little.