Re: Compressive images test

On 13.09.2013, at 18:40, Jason Grigsby <jason@cloudfour.com> wrote:

> It seems to me that we're looking for a combination of multiple compressive images on a page using older mobile hardware. And we should have the same page with standard images as a baseline to compare to.
> 
> The challenge I run into is figuring out what is a reasonable number of compressive images to include. I'm pretty sure that we could manufacture a page that doesn't work ("Oh hey, we put a hundred compressive images on a page and it crashed the browser on a Blackberry 5. I guess compressive images don't work. </shrug>").
> 
> But that wouldn't be a fair or realistic test. Ultimately, the only test that matters is for your page and your content on the devices that you're supporting. 
> 
> However, simply saying that doesn't give us any good sense of where the trouble spots may be. And more worrisome in my mind is that if compressive images really do cause problems with older devices and we don't offer any more guidance than you should check on a project by project basis, that people will use that as an excuse to ignore the impact on older devices.
> 
> So it seems we need some way to define where the edges of the technique are. Something that says, "If you use a couple on a page, you're good. But if you have a a dozen on the page, you're going to have problems." Or "Use it, but make sure you test on device A, B, and C because if there are issues, those devices are going to be the ones that encounter the problems first."

Yes, this is overall correct. I have achieved Safari on iOS crashing several times with big "panoramic" photos that were used compressive but TBH while this is a use-case you can't say compressive images don't work because of this.
I would really be interested in some statistics that clarify when they work and where not.
Problem is to measure the impact. AFAIK only Chrome provides such DevTool measurement tools, right? Or do you have something in your mind how this can be measured automatically?

-Anselm

> -Jason
> 
> 
> 
> On Fri, Sep 13, 2013 at 7:10 AM, Ben Callahan <ben@heysparkbox.com> wrote:
> Not sure exactly what the test structure would look like, but we'd be happy to help do this. It's something we've been wondering as well…
> 
> 
> 
> 
> -- 
> Ben Callahan
> Sparkbox, President
> 
> 123 Webster Street, Studio 2
> Dayton, OH 45402
> 937.401.0915
> www.seesparkbox.com
> 
> 
> 
> 
> -- 
> +1 (503) 290-1090 o | +1 (503) 502-7211 m | http://cloudfour.com