News archive

13/10/2017

SVGGraph 2.25

“Improve the 3D pie slices” has been on the top of my SVGGraph to-do list for quite a long time. Version 2.25 of SVGGraph improves the 3D pie slices.

Version 2.24 added the ability to draw partial pie graphs by setting the end_angle option, but it didn't look at all right for the 3D pie graphs. If the example below had been drawn using version 2.24 the flat inside faces would be missing and the gradient over the curved sides would continue through the gap. Not good.

Better 3D pie slices

The improvements to the 3D pie slice drawing code means that some of the other pie graph types can be drawn in 3D now as well, so there are these two new graph types available:

ExplodedPie3DGraph
PolarArea3DGraph

The PolarArea3DGraph is just the PolarAreaGraph with the 3D code added, but I have made some changes to the existing ExplodedPieGraph as well as adding the ExplodedPie3DGraph. The code for calculating the explosion for both is in a separate file now, with the direction calculated better (because I could see from the 3D version of the exploded pie graph that the slices could sometimes end up overlapping.) There isn't a 3D version of the DonutGraph yet, but it is on the to-do list (the internal curves make it a bit more tricky).

Read more . . .

21/08/2017

SVGGraph 2.24.1

SVGGraph 2.24.1 exists to release some changes I have made over the last few months while I have been concentrating on other projects. The main change in this version is the addition of a StackedBarAndLineGraph to go with the existing BarAndLineGraph:

StackedBarAndLineGraph

The StackedBarAndLineGraph supports all the usual options for stacked bar graphs and multi-line graphs, with one exception: the dataset_axis option can only be used to put line datasets on the second axis. This is because bars stacked on top of each other must use the same scale or the graph will not make sense.

Read more . . .

04/08/2017

TrayBlank 1.9

TrayBlank 1.9 actually adds something new for a change - options to start the screensaver directly. I've also updated the code that draws the tray icon to make it look clearer, and made the progress bar a lot more likely to be accurate.

The new option to start the screensaver directly is in a submenu called “Screensaver start” along with the existing delay option and another new option, “Directly after timeout”. What a direct start does differently is starting the screensaver as a new process instead of sending a message to Windows telling it to start the screensaver.

This means that the screensaver is run the same way as if you double-clicked on the screensaver icon itself. It will run on your main desktop instead of the empty desktop that Windows creates for screensavers to run on, so it has access to Windows Explorer and any other windows running there. Microsoft don't separate things out for no reason - if you don't trust your screensaver, don't run it directly! Actually, if you don't trust your screensaver you shouldn't run it at all. Uninstall it and run a virus and malware scan.

Read more . . .

12/07/2017

JPEG Saver 4.19.2

JPEG Saver 4.19.2 is another bugfix release. The main bug fixed in this version is a crash when using the “Desktop” or “Desktop Pile” backgrounds. I've also fixed the items editor not remembering the item name or its enabled/disabled state when creating a new item.

There are two new transitions in this version, “White Out” and “Colour Fade”. They both fade the current image out then fade the new image in, with “White Out” fading to white in between and “Colour Fade” fading to the colour in the centre of the first screen in between.

The new version is available from the downloads page.

28/06/2017

JPEG Saver 4.19.1

This shouldn't surprise anyone - the first new JPEG Saver release since I moved all the drawing code around and added in Direct2D and DirectWrite just fixes a couple of bugs. Not very exciting bugs, either.

The first bug was the image filename being forgotten when the image was being processed. This meant that some of the interactive options were failing in new and unusual ways (not just the usual ways that they fail because of the separate desktop that Windows runs screensavers on).

While I was fixing that one I discovered another bug - memory was being allocated for a copy of the image, which wasn't being used. This one didn't do any harm because the memory was released again soon after, but it might have slowed things down a bit.

Read more . . .

Next old news »