deniseconway

Hi Fellyph, it would appear that transient caching is actually not running and there is no option to enable it. It would seem that it isn’t on due3 to external object cache and it says I’m not using a persistent object cache (which I am, I use the Redis object cache).

The message that shows when I expand the transient cache object in site health is as follows:

Transient caching of parsed stylesheets is not used due to external object cache
AMP
On sites which have highly variable CSS and are not using a persistent object cache, the transient caching of parsed stylesheets may be automatically disabled in order to prevent a site from filling up its wp_options table with too many transients. Examples of highly variable CSS include dynamically-generated style rules with selectors referring to user IDs or elements being given randomized background colors. There are two filters which may be used to configure the CSS transient monitoring: amp_css_transient_monitoring_threshold and amp_css_transient_monitoring_sampling_range.

@fellyph Transient cache appears to be on. At least it doesn’t’ show as being off when I check site health. Is there somewhere else I can check it?

@fellyph and @westonruter
Further update…

I reduced the size of the image on the homepage to 17kb and I was still seeing the LCP issue, with the image still showing as the largest content drawn in the display. So, I removed the image and instead just created a cover block with the same text, and now it flags the text as the largest item drawn.

To me, this indicates that the image size is not the issue, especially given that the image has been shown all along as being the largest drawn element on the page, even when the LCP was less than 2.5sec.

Since the 5.9 update to wordpress, the LCP has not dropped below 2.5 sec. Taking away the image doesn’t help. I am now at a loss as to what else I can try. Any ideas?

@fellyph ignore my last comment, I have read a message further up from @westonruter explaining that my use of fixed background for the image prevents it from being responsive. I’m using the squoosh tool to make the images smaller and I will load them all again. Hopefully that sorts it.

Hi @fellyph,

Today I sat down to begin the work of resizing all my images, but when I check my content folder, I can see that the images are already there in all sizes (webp format) ranging in size from about 2kb for smallest size and up to about 80 kb for full size. I cannot figure out what to do to make wordpress serve the smaller images for the smaller screens. Do you happen to now where I can find an article on this? All of my searches are drawing a blank (mostly they point to plugins and themes but none for AMP).

I have already contacted my theme provider for assistance but they have said it’s a wordpress issue.

Also, the “mother thread” seems to be slowing things down:

Reduce the workload on the mother thread  2.1 s
Reduce the time it takes to interpret, compile and run JS code. It usually helps to reduce the size of the JS resources that are sent. Read moreTBT
Category
Time spent
Script Evaluation
923 ms
Style & Layout
525 ms
Other
386 ms
Rendering
145 ms
Press HTML & CSS
74 ms
Garbage Collection
16 ms
Script Parsing & Compilation
14 ms

How can that be reduced?

My google search impressions have taken a dive in the last week. I really need to find a resolution.

The site now has 0% of good URL’s for page experience.

What does this one indicate?

The order chains below show which resources are loaded with high priority. See if you can improve page load time by shortening the chains, reducing the size of downloaded resources, or postponing the download of unnecessary resources. Read more .FCPLCP
Maximum critical chain latency: 1,240 ms
First navigation
https://theflowerempowered.com
… Theflowerempowered.com / 01fd293a-9ad6-4029-80f7-af64ef09b6ea- 20 ms, 0.00 KiB

I’m so disappointed that wordpress did the autoupdate. There was definitely something in that update that 5.9 update that caused the slow down of the site as the hero image was showing up for LCP before but the site was not failing because of it.

What could have changed in 5.9 that I could potentially reverse?

Ok, here comes the dumb question for the day…. drum roll…
Where do I set the property for the image? Do I just add “data-hero” in the additional CSS classes box for the image?

Thanks!

Hi @fellyph, I ran my url through the amp/dev/page-experience checker and it has suggested that I use an amp optimizer to add the following script:

<code><span class="o"><</span><span class="n">link</span> <span class="k">as</span><span class="o">=</span><span class="n">script</span> <span class="n">href</span><span class="o">=</span><span class="n">https</span><span class="p">:</span><span class="o">//</span><span class="n">cdn</span><span class="o">.</span><span class="n">ampproject</span><span class="o">.</span><span class="n">org</span><span class="o">/</span><span class="n">v0</span><span class="o">.</span><span class="n">js</span> <span class="n">rel</span><span class="o">=</span><span class="nb">preload</span><span class="o">></span>
</code>

I can see that it lists the AMP plugin for WordPress as an AMP optimizer. Is it possible to add the suggested code? Or is it there already?

Otherwise, I could try adding a cloudflare worker but I have looked at that already and have no clue where to find the command line for setting it up.

Thanks.

I’m also looking in to serving different image sizes.

Looking at the LCP on pagespeed insights, it’s still long, even though the server response time is now small. I can see this message:

Avoid chaining critical requests 1 chain found
The Critical Request Chains below show you what resources are loaded with a high priority. Consider reducing the length of chains, reducing the download size of resources, or deferring the download of unnecessary resources to improve page load. Learn more.FCPLCP
Maximum critical path latency: 1,270 ms
Initial Navigation
https://theflowerempowered.com
https://theflowerempowered.com
…theflowerempowered.com/09c66e57-37ab-

Is there something I can do to speed that up?

Also, it’s flagging the hero image in the LCP suggestions. It’s already in WebP format. Is there something else I should do to make that load faster?

Thanks again for your continued support

So I have configured persistent object caching as well as having the transient caching available. I am now seeing fast server response times, but the LCP is still over 2.5 seconds. Any suggestions on what I can do to speed it up?

Thanks @westonruter, I’ve installed the plugin and re-enabled transient caching. Let’s see how things go ??

Fantastic!! I will await the workaround. Really appreciate your help ??

@fellyph checking the site health this morning and I notice that the transient cache is disabled again. Is there something that enables or disables it? Or is there something I can do to ensure it stays enabled?

@fellyph I don’t want to speak too soon but it looks like enabling transient caching has brought my score back to 99 on pagespeed insights. I’ve just initiated validation of the fix for LCP issue: longer than 2.5s (mobile) on Search Console.

I’ll reply back when the validation is complete. The page experience is currently showing 1.5% good pages, so hopefully we’ll see that rising back to 100% in the coming days!

I will keep you posted and thank you again for your persistence with this issue. Hopefully we can mark it as resolved within a few days ??