# P2D vs JAVA2D - strange differences in PGraphics methods and masking
**URL:** https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853
**Category:** Processing.py
**Created:** [June 5, 2019, 2:55am UTC](https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853 "2019-06-05T02:55:39Z")
**Posts on this page:** 3
**Page:** 2
### Author:  [@neilcsmith](https://discourse.processing.org/u/neilcsmith)
#### Post date: [June 7, 2019, 1:32pm UTC](https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853/21 "2019-06-07T13:32:04Z")
> [@GoToLoop](#):
>
> But it’s pretty useless given we need to use a `try {} / catch () {}` block for it, even though it’s merely a **get()** alias:
Yes, and breaks semantics for `clone()` - just pointing out that you don’t really want what you say you want! 😉
> [@GoToLoop](#):
>
> Here’s my attempt of cloning a PGraphics …
Yes, definitely broken in 3.5.3. In fact, looks like some other uses of the pixel array are broken too. Going to do a diff on the sources.
However, creating a copy of a PGraphics via the pixel array is a daft idea anyway, particularly if the graphics is on the GPU. Just do this, which works correctly across renderers, and will perform vastly better -
```auto
static final String RENDER = P3D, URL =
"https://" + "Forum.Processing.org/" + "processing-org.jpg";
PImage img;
PGraphics clonedPg;
void settings() {
img = loadImage(URL);
size(img.width, img.height, RENDER);
smooth(8);
noLoop();
}
void setup() {
// Buggy for all renderers but JAVA2D under PDE 3.5.3:
final PGraphics pg = createGraphics(width, height, RENDER);
pg.beginDraw();
pg.blendMode(REPLACE);
pg.image(img, 0, 0);
pg.endDraw();
clonedPg = clonePGraphics(pg);
}
void draw() {
image(clonedPg, 0, 0);
}
PGraphics clonePGraphics(final PGraphics pg) {
final int w = pg.width, h = pg.height;
final PGraphics cloned = createGraphics(w, h, RENDER);
cloned.beginDraw();
cloned.blendMode(REPLACE);
cloned.image(pg, 0, 0);
cloned.endDraw();
return cloned;
}
```
---
### Author:  [@GoToLoop](https://discourse.processing.org/u/GoToLoop)
#### Post date: [June 7, 2019, 4:07pm UTC](https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853/22 "2019-06-07T16:07:54Z")
> [@neilcsmith](#):
>
> `cloned.blendMode(REPLACE);`
You should set the **blendMode()** back to its default BLEND at the end: ⚠
> **[blendMode() / Reference](https://processing.org/reference/blendMode_.html)**
>
> Blends the pixels in the display window according to a defined mode. There is a choice of the following modes to blend the source pixels (A) with the ones of pixels already in the display window (B). …
```auto
cloned.beginDraw();
cloned.blendMode(REPLACE);
cloned.image(pg, 0, 0);
cloned.blendMode(BLEND);
cloned.endDraw();
```
Also, `cloned.image(pg, 0, 0);` unnecessarily stores the original PGraphics object _pg_ in _cloned_’s internal WeakHashMap. #️⃣
Therefore the _cloned_ PGraphics object isn’t fully detached from the original 1. 🙄
---
### Author:  [@neilcsmith](https://discourse.processing.org/u/neilcsmith)
#### Post date: [June 7, 2019, 4:27pm UTC](https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853/23 "2019-06-07T16:27:47Z")
> [@GoToLoop](#):
>
> You should set the **blendMode()** back to its default BLEND at the end:
Yes, good point!
> [@GoToLoop](#):
>
> Also, `cloned.image(pg, 0, 0);` unnecessarily stores the original PGraphics object _pg_ in _cloned_ ’s internal WeakHashMap.
a) who cares?!
b) no, it doesn’t! (well, with this renderer anyway) 😀
[Previous page](https://discourse.processing.org/t/p2d-vs-java2d-strange-differences-in-pgraphics-methods-and-masking/11853.md?page=1)