Speed up of Mondrian

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Speed up of Mondrian

abergel
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

Tudor Girba-2
Great news. What were the changes about?

On my machine (MacBookPro 2,6GHz i7) and in a Moose 4.8 image, I get:
- old version ==> 131966
- new version ==> 73591

Which is great.

However, when I do the same with the original Mondrian, I get:

[| view |
view := MOViewRenderer new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view root applyLayout ] timeToRun.

==> 40818

I would be curious to know where the difference comes from.

Cheers,
Doru



On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev



--

"Every thing has its own flow"

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

Usman Bhatti
In reply to this post by abergel
The numbers I got on my machine (MacBookPro 2,3GHz i7):


On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

==> 78380
With old Mondrian:
==> 42625

That's strange because for me one of the purposes of building Roassal was to improve performance but the numbers are suggesting contrary.


Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

Stéphane Ducasse
I imagine that Roassal is doing more.
But indeed this is interesting. 

The numbers I got on my machine (MacBookPro 2,3GHz i7):


On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

==> 78380
With old Mondrian:
==> 42625

That's strange because for me one of the purposes of building Roassal was to improve performance but the numbers are suggesting contrary.


Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

Tudor Girba-2
In reply to this post by Usman Bhatti
Hi,

Roassal was certainly not done out of performance reasons, but due to lack of flexibility in terms of animation :).

Doru

Sent from my iPhone

On Aug 9, 2013, at 9:43, Usman Bhatti <[hidden email]> wrote:

The numbers I got on my machine (MacBookPro 2,3GHz i7):


On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

==> 78380
With old Mondrian:
==> 42625

That's strange because for me one of the purposes of building Roassal was to improve performance but the numbers are suggesting contrary.


Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

Usman Bhatti
On Fri, Aug 9, 2013 at 2:20 PM, Tudor Girba <[hidden email]> wrote:
Hi,

Roassal was certainly not done out of performance reasons, but due to lack of flexibility in terms of animation :).

Ok, then I didn't understand it correctly :).
 

Doru

Sent from my iPhone

On Aug 9, 2013, at 9:43, Usman Bhatti <[hidden email]> wrote:

The numbers I got on my machine (MacBookPro 2,3GHz i7):


On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
Hi!

I've just integrated the fix of Mathieu.

1.395 : Integrated optimization from Mathieu. Excellent job!
[| view |
view := ROMondrianViewBuilder new.
view nodes: Object withAllSubclasses.
view edgesFrom: #superclass.
view treeLayout.
view applyLayout ] timeToRun
=> 91856 before this version
=> 48519 with this version

==> 78380
With old Mondrian:
==> 42625

That's strange because for me one of the purposes of building Roassal was to improve performance but the numbers are suggesting contrary.


Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev



_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: Speed up of Mondrian

abergel
In reply to this post by Tudor Girba-2
:-)
I did not think about comparing with the old Mondrian.
I will try.

Alexandre


On Aug 9, 2013, at 2:38 AM, Tudor Girba <[hidden email]> wrote:

> Great news. What were the changes about?
>
> On my machine (MacBookPro 2,6GHz i7) and in a Moose 4.8 image, I get:
> - old version ==> 131966
> - new version ==> 73591
>
> Which is great.
>
> However, when I do the same with the original Mondrian, I get:
>
> [| view |
> view := MOViewRenderer new.
> view nodes: Object withAllSubclasses.
> view edgesFrom: #superclass.
> view treeLayout.
> view root applyLayout ] timeToRun.
>
> ==> 40818
>
> I would be curious to know where the difference comes from.
>
> Cheers,
> Doru
>
>
>
> On Thu, Aug 8, 2013 at 9:32 PM, Alexandre Bergel <[hidden email]> wrote:
> Hi!
>
> I've just integrated the fix of Mathieu.
>
> 1.395 : Integrated optimization from Mathieu. Excellent job!
> [| view |
> view := ROMondrianViewBuilder new.
> view nodes: Object withAllSubclasses.
> view edgesFrom: #superclass.
> view treeLayout.
> view applyLayout ] timeToRun
> => 91856 before this version
> => 48519 with this version
>
> Cheers,
> Alexandre
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
>
>
> --
> www.tudorgirba.com
>
> "Every thing has its own flow"
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel  http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev