Lite speed cache

Hello,
Today i've made some adjustments to the settings of my Lite Speed Cache.

1. Lazy Load Images - ON
2. Responsive Placeholder - ON
3. Add Missing Sizes - ON

Till that moment the performance of the site according to PageSpeed Insights was with score 90 on mobile. Desktop was 99.
After that changes I've made the score downgraded to 59 on mobile but the desktop score is the same.
After my tests I can say that the site is faster but Google doesn't think so.
What do you think can cause the problem?
I don't know if I should get back to the old settings or stick to the new. In the meantime GTmetrix gives me 98% score and everything lights in green.
What do you think can cause the probem with PageSpeed Insights?
Thank you!

PageSpeed reports the issues below the score. Where are your orange/red areas? First Contentful Paint? Blocking Time? Large Contentful Paint?

PageSpeed is usually very theory oriented, while GTmetrix has more practical use. GTMetrix also breaks down everything that loads on your website and can show you scripts, plugins and so on that run and slow things down, if any at all.

it doesnt really matter what that tooll says, you can have 100 but if your real experience which shows in GSC in core web vitals and page experience are low then it's irelevant, many tools are optimized just for the lab scores but real life performance is weak. its a case of trail and error there is no just one setting, also are you ruinning the litespeed server with the tool or the tool alone? if its just thge tool then there are better options

There are so many options in the Litespeed cache plugin that it is difficult for a third party on a forum to diagnose someone else's problem.

Did you clear all caches?
What CSS and Javascript optimisation settings are you using? Many of these settings can cause conflict and break your site.
If you're using UCSS this cache needs to be cleared when making changes. It may also take a little while for the UCSS to be recreated on your CDN, so you should wait a few minutes before doing any tests.

Just in case anyone is not aware...

For the Lite Speed Cache to work, your server (host) VPS, Cloud, whatever... needs to be running Lite Speed itself.

Otherwise the plugin will not work ;)

Quote:

---

Originally Posted by AussieDave

Just in case anyone is not aware...

For the Lite Speed Cache to work, your server (host) VPS, Cloud, whatever... needs to be running Lite Speed itself.

Otherwise the plugin will not work ;)

---

Now it can run as a regular cache plugin but there are better otpions, to feel the beneift you need the server and even then it takes quite a while to perfect the correct speed for real time use

Quote:

---

Originally Posted by wonderpunter

to feel the beneift you need the server and even then it takes quite a while to perfect the correct speed for real time use

---

Really? I've been using it for a few years now, on my beefy, fully managed VPS. From my experience, if you can configure Yoast, or the plethora of WP plugins these days, then the lite-speed cache isn't a time vampire to setup, to obtain very impressive results.

Thing is though, from my 23 years affiliate exp, the one thing that shoots most affiliates in the foot, they scrimp of cheap-hosting, or oversold shared hosting. Would you run a business from a cheap ass tent, knocked up in your local park? NO.

Doesn't mean you have to go all out and blow 100's per month, but definitely find @ min, a shared hosting account which enables a decent share of server resources, and isn't over-crowded.

With all that said and done, the proof comes in pudding... Here's mine, where's yours?

https://www.gpwa.org/forum/images/im...dC/gmetrix.jpg

Of course, I've also tweaked theme coding, which also included highly reducing the database calls, and removing useless coding too.

I only use the barest min of WP plugins (5). Those plugins are AIOSEO, LS cache, Cookie accept notice, and a couple others which dictate functionality, that adding the hard code to functions.php wouldn't be worth the hassle.

Used to offer a speed up service to affiliates running WP sites, but just like cheap hosting, most don't want to shell out a few hundred.

I prefer to use google page speed score i have found gtmetrix and toolspingdom dont always correlate to real world speed figures in core vitals., with the mobile results it took me a lot of configuration to get this, I use good vps litespeed plus a couple of other things, Geotargeting and speed are pretty difficult but i have managed it i have quite a heavy site . You are right you need to spend money and in the top 10 thats what makes the difference soemtimes.
Desktop:
Attachment 8920
Mobile
Attachment 8918
Mobile core web vitals
Attachment 8919

Quote:

---

Originally Posted by wonderpunter

I prefer to use google page speed score i have found gtmetrix and toolspingdom dont always correlate to real world speed figures in core vitals., with the mobile results it took me a lot of configuration to get this, I use good vps litespeed plus a couple of other things, Geotargeting and speed are pretty difficult but i have managed it i have quite a heavy site . You are right you need to spend money and in the top 10 thats what makes the difference soemtimes.
Desktop:
Attachment 8920
Mobile
Attachment 8918
Mobile core web vitals
Attachment 8919

---

Click on the "See Calculator" link on your image. Check the values from the different page speed versions. Even if your website did not change, the weight of the different section did, and the score changed from there. PageSpeed is as dry theory crafting as it gets. A (20%) + B (15%) + ..... = 100%.

How does a math formula corelate to real life experience ? For example in LH v10, CLS is weighted 25%, but in the previous version 15%, so all of the sudden it matters that much more and if you were on the edge at around 90, you are now in the "Need Improvement" section without anything changing, but the formula.

the lab tests really dont mean much at all.. even with the weigting it's all just synthetic I go by coreweb vitals and page speed from GSC, actually i was caught up in the last boradcore and I suspect that had something to do with it as my pages were good on lab scores but horrendus in real time. I think on the next update I should regain some points if I do then i will create a thread about the importance of it. if not then will search for other ideas lol.