Skip to main content
Topic: Test Web Sites (Read 12416 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Test Web Sites

If you are bored and got nothing to do, test your sites or friends' sites or any sites you want here:

1. https://gtmetrix.com

2. http://www.webpagetest.org

Though web test like this may not be any use to you, it could be fun, especially if you make a screenshot and paste it as a reply in here.  :D


Re: Test Web Sites

Reply #2

I believe it is good except for the one you got F. You can ignore the one with X. You can tail them down to its details by clicking the F.

I think you can also get rid of that leverage browsing caching failure if you set proper expire in apache or nginx directive. Just google and you will find out about it.  ;)


Re: Test Web Sites

Reply #4

Google Pagespeed is another similar website.

https://developers.google.com/speed/pagespeed/insights/

Yes.  You can improve that by, for instance on Apache, adding something like this:

Code: [Select]
# BEGIN Expire headers
<ifModule mod_expires.c>
# Add correct content-type for fonts
# AddType application/vnd.ms-fontobject .eot
AddType application/x-font-ttf .ttf
AddType application/x-font-opentype .otf
AddType application/x-font-woff .woff
AddType image/svg+xml .svg

# Compress compressible fonts
#  AddOutputFilterByType DEFLATE application/x-font-ttf application/x-font-opentype image/svg+xml

ExpiresActive on

# Add a far future Expires header for fonts
# ExpiresByType application/vnd.ms-fontobject "access plus 2592000 seconds"
ExpiresByType application/x-font-ttf "access plus 2592000 seconds"
ExpiresByType application/x-font-opentype "access plus 2592000 seconds"
ExpiresByType application/x-font-woff "access plus 2592000 seconds"
ExpiresByType image/svg+xml "access plus 2592000 seconds"

ExpiresDefault "access plus 5 seconds"
ExpiresByType image/x-icon "access plus 2592000 seconds"
ExpiresByType image/jpeg "access plus 2592000 seconds"
ExpiresByType image/png "access plus 2592000 seconds"
ExpiresByType image/gif "access plus 2592000 seconds"
ExpiresByType application/x-shockwave-flash "access plus 2592000 seconds"
ExpiresByType text/css "access plus 604800 seconds"
ExpiresByType text/javascript "access plus 648000 seconds"
ExpiresByType application/javascript "access plus 648000 seconds"
ExpiresByType application/x-javascript "access plus 648000 seconds"
ExpiresByType text/html "access plus 600 seconds"
ExpiresByType application/xhtml+xml "access plus 600 seconds"
</ifModule>
# END Expire headers

I think I may have gotten the boilerplate for that from Google (as in Google's help I just linked to or something much like it, not by searching on Google :P)

Edit: forgot to add this link https://developers.google.com/speed/docs/insights/LeverageBrowserCaching

Re: Test Web Sites

Reply #5

Google pagespeed is even trickier than others.

For nginx leverage browsing caching, you can use this in your vhost:
Code: [Select]
        # serve static files directly
        location ~* ^.+\.(jpg|jpeg|gif|css|png|js|ico|html|xml|txt)$ {
                access_log        off;
                expires 365d;
        }

        location ~*  \.(pdf)$ {
                expires 30d;
        }

For those who are using ISPConfig and nginx, this can be set in nginx_vhost.conf.master in conf-custom folder, so that all sites will automatically use the same.



Re: Test Web Sites

Reply #8

I get a C score on GTmetrics & elkarte 1.0
https://gtmetrix.com/reports/www.skodaclub.it/5UI21Buw
if I check the home page the score is lower since I use scaled images

It says gzip compression is not enabled, but I'm pretty sure it is enabled
https://checkgzipcompression.com/?url=https%3A%2F%2Fwww.skodaclub.it%2Findex.php%3Faction%3Dforum

On Pingdom I get a B score https://tools.pingdom.com/#!/bj4XHC/https://www.skodaclub.it/index.php?action=forum
sorry for my bad english

Re: Test Web Sites

Reply #9

Based on the above, the only thing that I can see my server is lacking is the load time for its sites. I am not truly sure what affecting this but my guess is because I am using HDD instead of SSD.

I have disabled swap and remove its directory on both of my HDD and install preload to ensure all frequently used software will be using my 8gb ram instead but still I cannot get a faster load time.

So unless and until I have an SSD to be used as my server main drive, I guess I will have to be satisfied of what the server is offering, for now.

 

Re: Test Web Sites

Reply #10

Yep the load time is similar to mine, but my page size and the requests are triple
Fully Loaded Time 3.1s Fully Loaded Time 3.7s
Total Page Size 148KB Total Page Size  692KB
Requests 9 Requests 37
I use an ssd drive, with 2GB Ram from Linode
sorry for my bad english

Re: Test Web Sites

Reply #11

Quote from: ahrasis – Based on the above, the only thing that I can see my server is lacking is the load time for its sites. I am not truly sure what affecting this but my guess is because I am using HDD instead of SSD.

I have disabled swap and remove its directory on both of my HDD and install preload to ensure all frequently used software will be using my 8gb ram instead but still I cannot get a faster load time.

So unless and until I have an SSD to be used as my server main drive, I guess I will have to be satisfied of what the server is offering, for now.
Keep in mind that, assuming your server is located in Asia, testing against an American or European server might easily add half a second or more of extra "loading" time. Pingdom gives my site 360 ms from Sweden and 1.17 s from Dallas. I wouldn't be surprised if it'd go over 2 s from Asia with no change in server performance whatsoever.

Re: Test Web Sites

Reply #12

That's a good observance @Frenzie. I didn't think of that when I wrote earlier.

Quote from: radu81 – Yep the load time is similar to mine, but my page size and the requests are triple
Fully Loaded Time 3.1s Fully Loaded Time 3.7s
Total Page Size 148KB Total Page Size  692KB
Requests 9 Requests 37
I use an ssd drive, with 2GB Ram from Linode
Anyway, I do think the load time should be faster if we are using SSD, but I am not so sure how much more faster.

By the way, what is your processor @radu81? And are using your server for just one site?

And as a further todo for this, I will be checking how ISPConfig is managing this as well. It could have something to do with it or its settings for the site, though I doubt it.

Re: Test Web Sites

Reply #13

I usually do the GTmetrics tests using a server in Europe since my vps is located in London. In the link posted below I forgot to change the server to Europe

Quote from: ahrasis – By the way, what is your processor @radu81? And are using your server for just one site?
model name      : Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
cpu MHz        : 2499.996
cache size      : 4096 KB
I am hosting more than one website on this vps and  usually it uses 0,9 - 1,1 GB of 2 GB RAM, CPU used is about 10% - 25%
sorry for my bad english

Re: Test Web Sites

Reply #14

I always feel these tools are always to take with a grain of salt:
https://gtmetrix.com/reports/www.elkarte.net/9jVbbnZ4
Yes, images are not scaled, sure.
But we use avatars of different sizes (at least two, likely 3) in different pages, what would be the total size of all the scaled avatars? Of course it would be the size of the unscaled avatar (topic display) è the size of the scaled avatar.
In terms of caching, it is much more efficient to just serve 1 size of avatars and let the browser deal with the caching and re-use.

Also, the "optimize images": Optimize the following images to reduce their size by 11.3KiB (17% reduction).
Considering the page is 284kb, 11 is about 5%, not 17%. And... well, okay, really, 11 kb makes any difference nowadays?  When the average website has pages on the realm of 3MBs?

And for fun:
https://gtmetrix.com/reports/www.nytimes.com/2Es85Dwm
https://gtmetrix.com/reports/xenforo.com/KAjAlhF9
https://gtmetrix.com/reports/www.vbulletin.com/3B5w5zWO
https://gtmetrix.com/reports/invisioncommunity.com/1r7CP8MO

:D
Bugs creator.
Features destroyer.
Template killer.