[Lazarus] Testing Rapberry Pi 3 performance

classic Classic list List threaded Threaded
22 messages Options
12
Reply | Threaded
Open this post in threaded view
|

[Lazarus] Testing Rapberry Pi 3 performance

Anthony Walter
I recorded a very brief video of the performance of Raspberry Pi and Lazarus is quite usable.

Video here:

http://www.getlazarus.org/setup/raspberry/

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Graeme Geldenhuys-3
On 2016-03-23 04:56, Anthony Walter wrote:
> I recorded a very brief video of the performance of Raspberry Pi and
> Lazarus is quite usable.


For a computer priced at £22, that is incredible value for money! I
still have the original RPi - maybe it's time for an upgrade. The work
awesome as little micro-servers too. :)


Regards,
  - Graeme -



--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Mark Morgan Lloyd
Graeme Geldenhuys wrote:
> On 2016-03-23 04:56, Anthony Walter wrote:> I recorded a very brief video of the performance of Raspberry Pi and> Lazarus is quite usable.
>
> For a computer priced at £22, that is incredible value for money! Istill have the original RPi - maybe it's time for an upgrade. The workawesome as little micro-servers too. :)

The RPi2 is superior in almost every way. I've got an RPi3 here but
won't have time to exercise it for at least a few days.

The Raspberry Pi Zero can, I believe, be run as a USB slave
(specifically, USB OTG can be enabled by a kernel patch). That opens the
possibility of using it to implement special-purpose input peripherals:
has anybody used FPC etc. to play with USB Gadget programming?

http://hackaday.com/2016/03/02/hack-a-ps2-keyboard-onto-your-pi-zero/
http://isticktoit.net/?p=1383

--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk

[Opinions above are the author's, not those of his employers or colleagues]

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Graeme Geldenhuys-3
On 2016-03-23 10:22, Mark Morgan Lloyd wrote:
> The RPi2 is superior in almost every way. I've got an RPi3 here but
> won't have time to exercise it for at least a few days.

My RPi 1 Model B (the original) is still ticking nicely even after so
many years of usage. At one stage I used used it as a web server, then
as a media player for over two years, running 24/7. Now I use it to test
FPC+fpGUI builds for the ARM target.

>From what I've read online, the RPi2 and RPi3 have way more memory and
processing speed than mine. So that should be a nice welcome as a lite
desktop / development environment.

Regards,
  - Graeme -

--
fpGUI Toolkit - a cross-platform GUI toolkit using Free Pascal
http://fpgui.sourceforge.net/

My public PGP key:  http://tinyurl.com/graeme-pgp

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Mark Morgan Lloyd
Graeme Geldenhuys wrote:
> On 2016-03-23 10:22, Mark Morgan Lloyd wrote:
>> The RPi2 is superior in almost every way. I've got an RPi3 here but
>> won't have time to exercise it for at least a few days.
>
> My RPi 1 Model B (the original) is still ticking nicely even after so
> many years of usage. At one stage I used used it as a web server, then
> as a media player for over two years, running 24/7. Now I use it to test
> FPC+fpGUI builds for the ARM target.

The RPi1 I've got here wasn't reliable, but it could have been a thermal
runaway and I'm supposed to be testing it on a battery.

>>From what I've read online, the RPi2 and RPi3 have way more memory and
> processing speed than mine. So that should be a nice welcome as a lite
> desktop / development environment.

If we continue this it would be better in -other, but I've got a spare 2
and 3 here and would quite like to explore xdmx as we've already
discussed. Potentially, a cluster of those could replace my desktop
systems: I think they're entirely adequate for development.

--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk

[Opinions above are the author's, not those of his employers or colleagues]

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

David Taylor
In reply to this post by Mark Morgan Lloyd
On 23/03/2016 10:22, Mark Morgan Lloyd wrote:

> Graeme Geldenhuys wrote:
>> On 2016-03-23 04:56, Anthony Walter wrote:> I recorded a very brief
>> video of the performance of Raspberry Pi and> Lazarus is quite usable.
>>
>> For a computer priced at £22, that is incredible value for money!
>> Istill have the original RPi - maybe it's time for an upgrade. The
>> workawesome as little micro-servers too. :)
>
> The RPi2 is superior in almost every way. I've got an RPi3 here but
> won't have time to exercise it for at least a few days.
[]

The RPI-2 is superior to what?

Using the RPi-3 is a significantly more pleasant experience than the
RPi-2, even just headless when compiling things.  Highly recommended.  I
did have some minor issues with Jessie, with the serial port on the
RPi-3, and with gpsd not suto-starting.  I detailed those towards the
end of this page:

   http://www.satsignal.eu/ntp/Raspberry-Pi-quickstart.html

I've not had chance to try Lazarus, though.
--
David
Web: http://www.satsignal.eu


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael Schnell
On 03/23/2016 03:00 PM, David Taylor wrote:
>
>
> Using the RPi-3 is a significantly more pleasant experience than the
> RPi-2, even just headless when compiling things.  Highly recommended.  
> I did have some minor issues with Jessie, with the serial port on the
> RPi-3, and with gpsd not suto-starting.
But still no "official" 64 bit OS, yet :-(

-Michael


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

David Taylor
On 23/03/2016 14:18, Michael Schnell wrote:

> On 03/23/2016 03:00 PM, David Taylor wrote:
>>
>>
>> Using the RPi-3 is a significantly more pleasant experience than the
>> RPi-2, even just headless when compiling things.  Highly recommended.
>> I did have some minor issues with Jessie, with the serial port on the
>> RPi-3, and with gpsd not suto-starting.
> But still no "official" 64 bit OS, yet :-(
>
> -Michael
>
>
> --

But with "only" 1 GB of memory, 64-bit may be no better than 32-bit.
The hardware, though, is 64-bit with the advantages that brings in
memory access speeds.  I've seen nothing of the Raspberry Pi cards which
I use which would benefit from a 64-bit version.

--
David
Web: http://www.satsignal.eu


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael Schnell
On 03/23/2016 03:32 PM, David Taylor wrote:
>
> But with "only" 1 GB of memory, 64-bit may be no better than 32-bit.
> The hardware, though, is 64-bit with the advantages that brings in
> memory access speeds.  I've seen nothing of the Raspberry Pi cards
> which I use which would benefit from a 64-bit version.
>
I do know that with X86 the memory efficiency with 64 is a lot worse
than with X32 and might result in speed degradation due to worse cache
usage. I have no knowledge if this is similar with aarch64.

-Michael

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Bo Berglund
In reply to this post by David Taylor
On Wed, 23 Mar 2016 14:00:57 +0000, David Taylor
<[hidden email]> wrote:

>I've not had chance to try Lazarus, though.

You might want to look at my installation script for installing FPC
3.0.0 and Lazarus 1.6 on a bare newly started RPi3 with Raspbian
Jessie:
http://blog.boberglund.com/install_laz_pi.sh

Revisions to download are modified in the script configuration at the
top. It needs a user response when it is running the install of the
seed compiler, otherwise it is unattended.


--
Bo Berglund
Developer in Sweden


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Graeme Geldenhuys-3
On 2016-03-25 09:33, Bo Berglund wrote:
> installing FPC
> 3.0.0 and Lazarus 1.6

A word of caution, there is a major bug in FPC 3.0 regarding floating
point values (see the FPC mailing list), so I wouldn't recommend anybody
upgrade to FPC 3.0 at this point - at least until 3.0.2 is out. This
issue has been fixed in FPC Trunk already though.

Regards,
  - Graeme -

--
fpGUI Toolkit - a cross-platform GUI toolkit using Free Pascal
http://fpgui.sourceforge.net/

My public PGP key:  http://tinyurl.com/graeme-pgp

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael Schnell
On 03/25/2016 12:17 PM, Graeme Geldenhuys wrote:
> A word of caution, there is a major bug in FPC 3.0 regarding floating
> point values (see the FPC mailing list), so I wouldn't recommend
> anybody upgrade to FPC 3.0 at this point - at least until 3.0.2 is
> out. This issue has been fixed in FPC Trunk already though.
They say: "This happens only on Win64 with FPC 3.0 "

-Michael

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael Van Canneyt


On Tue, 29 Mar 2016, Michael Schnell wrote:

> On 03/25/2016 12:17 PM, Graeme Geldenhuys wrote:
>> A word of caution, there is a major bug in FPC 3.0 regarding floating point
>> values (see the FPC mailing list), so I wouldn't recommend anybody upgrade
>> to FPC 3.0 at this point - at least until 3.0.2 is out. This issue has been
>> fixed in FPC Trunk already though.
> They say: "This happens only on Win64 with FPC 3.0 "

No, it will also happen on 64-bit arm.

Michael.

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Mark Morgan Lloyd
In reply to this post by Michael Schnell
Michael Schnell wrote:
> On 03/25/2016 12:17 PM, Graeme Geldenhuys wrote:
>> A word of caution, there is a major bug in FPC 3.0 regarding floating
>> point values (see the FPC mailing list), so I wouldn't recommend
>> anybody upgrade to FPC 3.0 at this point - at least until 3.0.2 is
>> out. This issue has been fixed in FPC Trunk already though.

> They say: "This happens only on Win64 with FPC 3.0 "

Happens on all sorts of things, depending on what selection has been
made for certain underlying types. /Specifically/ happens on an RPi2, as
discussed in the bug report.

--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk

[Opinions above are the author's, not those of his employers or colleagues]

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Mark Morgan Lloyd
In reply to this post by Michael Van Canneyt
Michael Van Canneyt wrote:

> On Tue, 29 Mar 2016, Michael Schnell wrote:
>
>> On 03/25/2016 12:17 PM, Graeme Geldenhuys wrote:
>>> A word of caution, there is a major bug in FPC 3.0 regarding floating
>>> point values (see the FPC mailing list), so I wouldn't recommend
>>> anybody upgrade to FPC 3.0 at this point - at least until 3.0.2 is
>>> out. This issue has been fixed in FPC Trunk already though.
>> They say: "This happens only on Win64 with FPC 3.0 "
>
> No, it will also happen on 64-bit arm.

Or 32-bit on an RPi2 (Debian "Jessie" rather than Raspbian etc.).

--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk

[Opinions above are the author's, not those of his employers or colleagues]

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

[Lazarus] Feature Request: Insert {codepage UTF8} per default

Michael W. Vogel
Hi,

I'm thinking about the thread here
http://forum.lazarus.freepascal.org/index.php/topic,31939.msg206688.html#msg206688 
where Bart wrote "... and adding explicit codepage identifiers to
sourcecode that has string constants with characters that er not ASCII
(or ad -FcUTF8 to your default compiler options to achieve the same) ..."

I think he is right.

Since Lazarus makes per default the switch to UTF-8 encoded strings, it
would be good to hint the developer, that they use such encoded strings.

In the most cases the string magic works without a defined {$codepage
utf8}, but not if you want to assign a const to a Predefined String or
Unicodestring.

I add a project that demonstrate, that the assigning of strings with
{$codepage utf8} works better than without it.

I would prefer to set {$codepage utf8} per default, if a new
unit/project is created. I like it more than the general switch -FcUTF8,
cause each developer see the current codepage and you can use units with
other codepages too (what in reality probably nobody would do).

I add a possible patch, how I think, it could be.


Offtopic: In the added project: Why is a const 'abc' with {$codepage
UTF8} a Unicodestring (Windows7, 64bit, Lazarus 1.7 r52077M FPC 3.1.1
i386-win32-win32/win64 on FPC 3.1.1 r33371)?


Regards

Michl


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus

codepages.patch (10K) Download Attachment
StringAssignTest.zip (9K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael W. Vogel
In reply to this post by Mark Morgan Lloyd
Sorry, my reply enters the wrong place.

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Bo Berglund
In reply to this post by Graeme Geldenhuys-3
On Fri, 25 Mar 2016 11:17:36 +0000, Graeme Geldenhuys
<[hidden email]> wrote:
>A word of caution, there is a major bug in FPC 3.0 regarding floating
>point values (see the FPC mailing list),

You do not happen to know the subject of that thread on the FPC list?
I am using GMANE News mirror but I cannot find it without some hint on
the actual subject text.
I am working on a project where I will have to process floating point
values and do endian conversion and such stuff. So I am really anxious
that this could be affected by such a bug.

--
Bo Berglund
Developer in Sweden


--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Graeme Geldenhuys-3
On 2016-04-05 20:55, Bo Berglund wrote:
> You do not happen to know the subject of that thread on the FPC
> list?

It seems to be eluding me too. :-) Maybe post a new message and ask for
the Mantis bug ID to get more information.

All I remember was that it was a serious bug in currency handling on
win64 (and affected other 64-bit targets too), which was a show-stopper
for my client upgrading.

Regards,
  - Graeme -

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Reply | Threaded
Open this post in threaded view
|

Re: [Lazarus] Testing Rapberry Pi 3 performance

Michael Schnell
In reply to this post by Bo Berglund
[fpc-devel] Bug 29760 on FPC 3.0 Win64

-Michael

--
_______________________________________________
Lazarus mailing list
[hidden email]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
12