Comments on: Using a Mac mini for development

By: cubec

cubec — Fri, 15 Jan 2010 19:43:32 +0000

Great article, which is exactly what I’m looking for ), the power of google.

BTW, you can use Matrox for dual monitor with any vga port.

I’m setting up a small office and this is exactly what I’ve been looking at since I can’t afford to spend for 2 mac book pros.

By: Andy Brice

Andy Brice — Sun, 13 Sep 2009 20:53:02 +0000

>Can I use the DVD-D connector with the DVI-I cinema display?

I don’t remember the gory details. I suggest you ask your local Apple dealer.

By: Danny

Danny — Sun, 13 Sep 2009 18:28:24 +0000

Hey I just saw your article. I use a previous-gen Mini as my main dev machine. It has the DVI-connector which I use with an older cinema display. I’m going to upgrade to the newest mini and I assumed that the supplied DVI adaptor would work work with my display.

Is that not the case? Can I use the DVD-D connector with the DVI-I cinema display?

By: Andy Brice

Andy Brice — Sun, 21 Jun 2009 07:18:04 +0000

>QCoreApplication

“This class is used by non-GUI applications to provide their event loop. … QCoreApplication contains the main event loop, where all events from the operating system (e.g., timer and network events) and other sources are processed and dispatched. It also handles the application’s initialization and finalization, as well as system-wide and application-wide settings.”

NB/ You don’t have to use QCoreApplication to create a Qt console app.

By: Andy Brice

Andy Brice — Sun, 21 Jun 2009 07:12:52 +0000

>Is there any sort of KVM switch or similar that can handle dual-monitors

Not that I am aware of.

>and can the Mac feed two monitors?

Not that I am aware of.

By: Oleg

Oleg — Sat, 20 Jun 2009 11:04:16 +0000

Hi Andy,
One more thing to add – I found that their last wizard has check box for precompiled headers generation. Everything works OK. And question – what is the purpose of creating instance of QCoreApplication and running loop in console app?
QCoreApplication a(argc, argv);
return a.exec();
Thank you Andy that you growing my curiosity regarding QT and finally I try it out myself.

By: Alver

Alver — Sat, 20 Jun 2009 10:16:21 +0000

Thanks for the great article, and many useful follow-up comments. I’m in exactly the same position – I’m about to start porting a Windows app to Mac using a Mac Mini. One question: My windows setup has two DVI monitors, connected to a dual-DVI graphics card in the PC. I’d love to use the same two screens, as well as the same keyboard and mouse, on the Mac. Is there any sort of KVM switch or similar that can handle dual-monitors, and can the Mac feed two monitors?

By: cencio

cencio — Fri, 19 Jun 2009 10:26:44 +0000

I have been using Mac mini for developmment. I can not say that I`m satisfied with it. We bought mini`s at the end of 2007. Memory was upgraded up to 2 Gb, so it is acceptable. But Hdd is small and it is 5200 and it is very hard to upgrade it due to mini design. So performance is not good. imho it is easy and cheaper buy laptop, upgrade it`s memory + hdd and use windows/linux for development. mac mini is an affordable solution for mac specific (MacOS, IPhone)development tasks only.

By: Moof

Moof — Thu, 18 Jun 2009 22:26:21 +0000

distcc is your friend.

I have a mini under my table and the serious juice out where I don’t have to hear it or look at it.

By: Oleg

Oleg — Thu, 18 Jun 2009 14:15:33 +0000

I am glad if it will save you some time,
QT still building on my ancient machine, meanwhile I have some suggestions that might help you either.
I read the documentation now and it looks like code example not very correct there (at least for VC compiler it will not reduce the time of compilation. Once again I am not sure here – you have to try it yourself.

I think that correct way would be:

preheaders.h – where you have to include all qt headers
preheaders.cpp – I think this file must exist either to generate correct pch.

Then typical class in your application should look like
/// some.h
class Some
{
void foo();
}

/// some.cpp
#include “preheaders.h”
#include “some.h”
void Some::foo()
{
}

I do not think you change QT headers very often. The simplest way to understand the concept is generating some MFC or ATL application and look at their structure. (stdafx.h and stdafx.cpp)

/Gm – is very good thing but sometimes you have to rebuild whole application to fix some strange things if flag enabled.
As far as I know in VC 80 also exists flag that may enable utilizing the power of 2 cores. I never was interested in it though