I’ve just setup two pages on my website as a test.
Page one has two progressive JPG’s placed separately on a page.
Page two has the same two progressive images, but placed within a slider which flips between the two images (Cyclone Slider 2 for Wordpress).
I ran the test to see how much slower it would be using the slider, but have found some interesting results:
Page one (two separate images)
http://www.webpagetest.org/result/140811_G4_11M2/
Page two (the same images in a slider)
http://www.webpagetest.org/result/140811_04_11M3/
Results:
- The slider started rendering quicker.
- Document Complete was a lot quicker with a slider.
- The fully loaded time was slightly slower however with a slider.
On the repeat view though, the slider was actually quicker than the separate images for both document complete and fully loaded (but slower to start rendering).
This isn’t something I was really expecting and has convinced me to keep the slider I’m planning on my next website.
The only thing slightly puzzling about these results are that when using a slider, it seems the testing tool doesn’t recognise that they are progressive JPG’s and so marks down for that. My theory is that this is being caused by the slider possibly not displaying until the images are completely loaded.
If this is the case, I’m guessing it’s worth not using progressive JPG’s in the sliders as the progressive nature does add some file-size, for no benefit it seems?