> Can anyone tell me what is the optimum page speed? Are there any > guidance notes as in above a certain threshold is OK etc.
> Cheers
> -- > You received this message because you are subscribed to the Google Groups > "page-speed-discuss" group. > To post to this group, send email to page-speed-discuss@googlegroups.com. > To unsubscribe from this group, send email to > page-speed-discuss+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/page-speed-discuss?hl=en.
-- met hartelijke groeten/kind regards harry van der Horst M 0031643016999
> What is your current page speed and is your goal to have your site > load fast or to have a high score?
> On Apr 3, 1:58 pm, marketingmast <marketingst...@mastgrp.com> wrote: > > Hi All,
> > Can anyone tell me what is the optimum page speed? Are there any > > guidance notes as in above a certain threshold is OK etc.
> > Cheers
> -- > You received this message because you are subscribed to the Google Groups > "page-speed-discuss" group. > To post to this group, send email to page-speed-discuss@googlegroups.com. > To unsubscribe from this group, send email to > page-speed-discuss+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/page-speed-discuss?hl=en.
In my opinion there are 3 elements to page speed load time
1) DNS resolution time. This should be very fast. Most of the time it
should take less than 100milliseconds.
2) TCP connection time. This can vary but, should still be fast. I
would look for less than .5 seconds.
3) HTTP or HTTPS download time. This can vary according to the amount
of data on the page of course. It will also increase if you are using
https, and verification from Verisign. For consistent results use just
the time it takes to download the html portion of the page. Plain http
should be be at most 1 second. Https can take a bit longer but
shouldn't be more than 3 seconds.
FWIW, the items you are listing below usually make up less than 10% of the time for actually loading a page and are the components for just delivering the HTML (or individually for any of the front-end requests). The average page from the Alexa top 100k has 84 separate requests (83 more after the HTML is delivered) to actually build the page for the users: http://httparchive.org/trends.php#bytesTotal&reqTotal <http://httparchive.org/trends.php#bytesTotal&reqTotal>
If you are using Google Analytics then you can get a bunch of rich information about the actual performance your end users are seeing under Content->Site Speed. If you go to Page Timings and then select the "Performance" tab you can get histograms for the overall page load times as well as granular timings for the base HTML (including the DNS, socket connect and server response times). You can also graph the averages by region (if you don't have a distributed infrastructure then the performance will degrade as users get further from the server).
To answer the original question, there really isn't an "optimum". Every study I have seen shows continued positive gains and the point of diminishing returns will vary by site. Google's webmaster tools has some information under Labs->Site performance that shows you how your site is performing relative to all sites (from real users). Looks like the 20th percentile is around 1.5 seconds right now.
> In my opinion there are 3 elements to page speed load time
> 1) DNS resolution time. This should be very fast. Most of the time it
> should take less than 100milliseconds.
> 2) TCP connection time. This can vary but, should still be fast. I
> would look for less than .5 seconds.
> 3) HTTP or HTTPS download time. This can vary according to the amount
> of data on the page of course. It will also increase if you are using
> https, and verification from Verisign. For consistent results use just
> the time it takes to download the html portion of the page. Plain http
> should be be at most 1 second. Https can take a bit longer but
> shouldn't be more than 3 seconds.
Very good point. That's why Google is smart to keep their pages
simple. They just load faster that way. There are so many cluttered
websites that take forever to load, to the point where some just never
load correctly.
Your best bet is to keep your website simple, with graphics loading
locally, or generated on the fly (such as .cgi embedded in img tags).
Also keep offsite content, such as ads to a minimum. I know people
like to stack their website with ads and flash content, but it's just
frustrating for the users.
On May 18, 11:25 am, Pat Meenan <patmee...@gmail.com> wrote:
> FWIW, the items you are listing below usually make up less than 10% of
> the time for actually loading a page and are the components for just
> delivering the HTML (or individually for any of the front-end
> requests). The average page from the Alexa top 100k has 84 separate
> requests (83 more after the HTML is delivered) to actually build the
> page for the users:http://httparchive.org/trends.php#bytesTotal&reqTotal > <http://httparchive.org/trends.php#bytesTotal&reqTotal>
> If you are using Google Analytics then you can get a bunch of rich
> information about the actual performance your end users are seeing under
> Content->Site Speed. If you go to Page Timings and then select the
> "Performance" tab you can get histograms for the overall page load times
> as well as granular timings for the base HTML (including the DNS, socket
> connect and server response times). You can also graph the averages by
> region (if you don't have a distributed infrastructure then the
> performance will degrade as users get further from the server).
> To answer the original question, there really isn't an "optimum". Every
> study I have seen shows continued positive gains and the point of
> diminishing returns will vary by site. Google's webmaster tools has
> some information under Labs->Site performance that shows you how your
> site is performing relative to all sites (from real users). Looks like
> the 20th percentile is around 1.5 seconds right now.
> Thanks,
> -Pat
> On 5/18/2012 10:10 AM, elassoto wrote:
> > In my opinion there are 3 elements to page speed load time
> > 1) DNS resolution time. This should be very fast. Most of the time it
> > should take less than 100milliseconds.
> > 2) TCP connection time. This can vary but, should still be fast. I
> > would look for less than .5 seconds.
> > 3) HTTP or HTTPS download time. This can vary according to the amount
> > of data on the page of course. It will also increase if you are using
> > https, and verification from Verisign. For consistent results use just
> > the time it takes to download the html portion of the page. Plain http
> > should be be at most 1 second. Https can take a bit longer but
> > shouldn't be more than 3 seconds.
Don't load images thru your serverside script - it's a complete waste
of resources and will slow your server to the point of not being able
to serve requests.
On May 21, 4:08 pm, elassoto <elass...@gmail.com> wrote:
> Very good point. That's why Google is smart to keep their pages
> simple. They just load faster that way. There are so many cluttered
> websites that take forever to load, to the point where some just never
> load correctly.
> Your best bet is to keep your website simple, with graphics loading
> locally, or generated on the fly (such as .cgi embedded in img tags).
> Also keep offsite content, such as ads to a minimum. I know people
> like to stack their website with ads and flash content, but it's just
> frustrating for the users.
> On May 18, 11:25 am, Pat Meenan <patmee...@gmail.com> wrote:
> > FWIW, the items you are listing below usually make up less than 10% of
> > the time for actually loading a page and are the components for just
> > delivering the HTML (or individually for any of the front-end
> > requests). The average page from the Alexa top 100k has 84 separate
> > requests (83 more after the HTML is delivered) to actually build the
> > page for the users:http://httparchive.org/trends.php#bytesTotal&reqTotal > > <http://httparchive.org/trends.php#bytesTotal&reqTotal>
> > If you are using Google Analytics then you can get a bunch of rich
> > information about the actual performance your end users are seeing under
> > Content->Site Speed. If you go to Page Timings and then select the
> > "Performance" tab you can get histograms for the overall page load times
> > as well as granular timings for the base HTML (including the DNS, socket
> > connect and server response times). You can also graph the averages by
> > region (if you don't have a distributed infrastructure then the
> > performance will degrade as users get further from the server).
> > To answer the original question, there really isn't an "optimum". Every
> > study I have seen shows continued positive gains and the point of
> > diminishing returns will vary by site. Google's webmaster tools has
> > some information under Labs->Site performance that shows you how your
> > site is performing relative to all sites (from real users). Looks like
> > the 20th percentile is around 1.5 seconds right now.
> > Thanks,
> > -Pat
> > On 5/18/2012 10:10 AM, elassoto wrote:
> > > In my opinion there are 3 elements to page speed load time
> > > 1) DNS resolution time. This should be very fast. Most of the time it
> > > should take less than 100milliseconds.
> > > 2) TCP connection time. This can vary but, should still be fast. I
> > > would look for less than .5 seconds.
> > > 3) HTTP or HTTPS download time. This can vary according to the amount
> > > of data on the page of course. It will also increase if you are using
> > > https, and verification from Verisign. For consistent results use just
> > > the time it takes to download the html portion of the page. Plain http
> > > should be be at most 1 second. Https can take a bit longer but
> > > shouldn't be more than 3 seconds.
> > > The speed checker athttp://www.mysitespeed.com/speed/speedget.php > > > incorporates these metrics and will instantly tell you which elements
> > > are slower than a threshold.
> > > On Apr 3, 8:58 am, marketingmast<marketingst...@mastgrp.com> wrote:
> > >> Hi All,
> > >> Can anyone tell me what is the optimum page speed? Are there any
> > >> guidance notes as in above a certain threshold is OK etc.
I found that when I generated an image (such as an .rrd) and wrote it to disk, and then loaded it into the webpage it was much slower than generating graphics with .cgi embedded into img tags.
On Monday, May 21, 2012 11:43:26 AM UTC-4, Alex Mcauley wrote: > FWIW:
> Don't load images thru your serverside script - it's a complete waste > of resources and will slow your server to the point of not being able > to serve requests.
> On May 21, 4:08 pm, elassoto <elass...@gmail.com> wrote: > > Very good point. That's why Google is smart to keep their pages > > simple. They just load faster that way. There are so many cluttered > > websites that take forever to load, to the point where some just never > > load correctly.
> > Your best bet is to keep your website simple, with graphics loading > > locally, or generated on the fly (such as .cgi embedded in img tags). > > Also keep offsite content, such as ads to a minimum. I know people > > like to stack their website with ads and flash content, but it's just > > frustrating for the users.
> > On May 18, 11:25 am, Pat Meenan <patmee...@gmail.com> wrote:
> > > FWIW, the items you are listing below usually make up less than 10% of > > > the time for actually loading a page and are the components for just > > > delivering the HTML (or individually for any of the front-end > > > requests). The average page from the Alexa top 100k has 84 separate > > > requests (83 more after the HTML is delivered) to actually build the > > > page for the users: > http://httparchive.org/trends.php#bytesTotal&reqTotal > > > <http://httparchive.org/trends.php#bytesTotal&reqTotal>
> > > If you are using Google Analytics then you can get a bunch of rich > > > information about the actual performance your end users are seeing > under > > > Content->Site Speed. If you go to Page Timings and then select the > > > "Performance" tab you can get histograms for the overall page load > times > > > as well as granular timings for the base HTML (including the DNS, > socket > > > connect and server response times). You can also graph the averages > by > > > region (if you don't have a distributed infrastructure then the > > > performance will degrade as users get further from the server).
> > > To answer the original question, there really isn't an "optimum". > Every > > > study I have seen shows continued positive gains and the point of > > > diminishing returns will vary by site. Google's webmaster tools has > > > some information under Labs->Site performance that shows you how your > > > site is performing relative to all sites (from real users). Looks > like > > > the 20th percentile is around 1.5 seconds right now.
> > > Thanks,
> > > -Pat
> > > On 5/18/2012 10:10 AM, elassoto wrote:
> > > > In my opinion there are 3 elements to page speed load time
> > > > 1) DNS resolution time. This should be very fast. Most of the time > it > > > > should take less than 100milliseconds. > > > > 2) TCP connection time. This can vary but, should still be fast. I > > > > would look for less than .5 seconds. > > > > 3) HTTP or HTTPS download time. This can vary according to the > amount > > > > of data on the page of course. It will also increase if you are > using > > > > https, and verification from Verisign. For consistent results use > just > > > > the time it takes to download the html portion of the page. Plain > http > > > > should be be at most 1 second. Https can take a bit longer but > > > > shouldn't be more than 3 seconds.
> > > > The speed checker athttp://www.mysitespeed.com/speed/speedget.php > > > > incorporates these metrics and will instantly tell you which > elements > > > > are slower than a threshold.
> > > > On Apr 3, 8:58 am, marketingmast<marketingst...@mastgrp.com> > wrote: > > > >> Hi All,
> > > >> Can anyone tell me what is the optimum page speed? Are there any > > > >> guidance notes as in above a certain threshold is OK etc.
