Introduction

The packet wanted to change the size of an image. (You might notice the result in a previous post.) As always, we over-complicated the task. You see, rather than settling for just any editor, we here at drunkenpacket wanted an image editor that was satisfying.

So, we tried some out. And will give you the benefit of our opinion. Even though you didn’t ask. The basis of comparison was some sort of non-objective sense of snappiness and responsiveness of dragging a selected segment of an image. The image in question was chosen to be representative of what would be produced by a reasonable mid-range digital camera: a 2304 by 3456 pixel JPEG. (No, we’ll not show you the image. Use your imagination.)

We selected a large chunk of it, made this a new layer and panned around the image at full size as one might be wont to do when looking for all of the dust particles thoughtfully included by the scanner.

The Candidates

• Imagewell
• Pixelmator
• Iris
• Acorn
• Seashore
• Shake

There has been (relatively recently) a spate of new Macintosh image editors. Others have written serious reviews of them. This is not a serious review. But it might suggest some novel alternatives.

Imagewell

We had previously used a free version of Imagewell for all our blog-related image twiddling. (Like removing the parts we didn’t want you see. No, not the naughty bits. More like our thumbs bulging enormously into the phone camera frame.)

Imagewell does what we need for DP image processing: some simple pan, resize, crop etc. Except that it use to be free and now it isn’t. The packet finds himself oddly annoyed by this sort of infinite inflation. So, we skipped it.

Pixelmator

The packet noted how the marketing contents for Pixelmator make prominent use of supermodels (or at least young women pretending to be) in its sample screen shots. This sums up everything wrong with Pixelmator. It’s all about looking beautiful and not necessarily about being particularly useful.

In our simple test, while dragging a rectangular layer proceeded smoothly, (That would be CoreGraphics for you perhaps.) the non-rectangular selection drag chugged along. The packet refuses to pay for image processing software today that is not as fast as MacPaint on his Mac SE.

Purists may blither on about how this is an unfair comparison: the image of today is at least 4bytes/pixel instead of 1b. Pish posh: Core duo 2 can sustain around 8 billion instructions/second vs the 68000’s paltry 4 million and I have a GPU. My drags should go fast. Always.

Iris

We were tired and wanted more coffee so we looked at the website, said isn’t that nice and closed the window.

Acorn

We so wanted to like Acorn. It has such a rational interface compared to Pixelmator. The author has really tried to build a better paint program. In terms of usability, Acorn is the best. One can even write plugins (in Python no less and the reader already knows that the packet has a fondness for snakes.)

Only one thing… Acorn was slower than Pixelmator. The end of a drag left the CPU meter pegged at a 100% for at least 5 seconds.

Shake

We’ll talk about using Shake as a paint program again. And have before. Shake is way more powerful than all of these other programs. It drags the layers plenty fast. But even the packet in a dervish fit of quixotic tool taste would be willing to admit that it is the one package here that doesn’t fit with the others. It would take longer for the packet to remember how to import an image into Shake than the the entire task would take in a more normal piece of software.

Seashore

We had looked at Seashore before. The website does not impress. The previous version of the software had seemed, well, sparse. (Especially after the baroque glitz of Pixelmator.)

But hey, the packet is open minded. Sometimes.

Seashore is open source. Free/Libre is good on financial and ideological grounds. And it’s not too shabby looking either. And fast enough. Faster than the others. Image drags were smooth. Hmm. Perhaps there’s something to be said about a strong underlying core image representation as a set of tiles? Perhaps the packet should contribute.

Conclusion

Seashore for our blog-related image fix-ups. And maybe we’ll get out our coding beach towel.