How to achieve lower TTFB?

ChievoChievo OG
in Technical

I have a dedicated server with hetzner in Falkenstein(nginx+fast cgi ) +cdn free cloudflare and dns with cloudflare .
Basically the ttfb for europe is OK around average 120ms . Meanwhile i have average 480ms for the USA and 620 ms for ASIA and 910ms for Australia.

on the another hand rocket.net have 57ms for europe ttfb. 53ms for usa and 185 for asia and 43 for australia.

Which are their secret combination? What do you think?And how could be possible to improve my ttfbs . Thanks

I believe in the good luck. Harder than I work luckier i get.

«1

Comments

  • skhronskhron Hosting Provider
    edited February 17

    Maybe there is buffering issue in your current configuration? 910ms TTFB is way too high.

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:
    Maybe there is buffering issue in your current configuration? 910ms TTFB is way too high.

    Could be the case.How could be possible to investigate it ?Any advice or guide or what to check would be highly appreciated. I mean 910ttfb for australia meanwhile for some points of europe is on the 70s the same configuration.

    I believe in the good luck. Harder than I work luckier i get.

  • skhronskhron Hosting Provider

    @Chievo, please check this

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:
    @Chievo, please check this

    many thanks ! it is on like on the default version. the server is not overcrowded etc

    I believe in the good luck. Harder than I work luckier i get.

  • skhronskhron Hosting Provider

    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I believe in the good luck. Harder than I work luckier i get.

  • skhronskhron Hosting Provider

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    In theory with paid plans you have access to more POPs and better speed . In theory at least from what i have read in different websites not sure at 100%

    I believe in the good luck. Harder than I work luckier i get.

  • skhronskhron Hosting Provider

    @Chievo said:

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    In theory with paid plans you have access to more POPs and better speed . In theory at least from what i have read in different websites not sure at 100%

    Yeah but you don't need 100% of Cloudflare PoPs to lower your TTFB for Australia from 980ms to 440ms at least (already x2 faster!)

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:

    @Chievo said:

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    In theory with paid plans you have access to more POPs and better speed . In theory at least from what i have read in different websites not sure at 100%

    Yeah but you don't need 100% of Cloudflare PoPs to lower your TTFB for Australia from 980ms to 440ms at least (already x2 faster!)

    I would try bunnycdn as well . Yeah pro plans are going to open some pops not all of them :( I understand your point but the reality is that unfortunately. Let s try different solutions in order to obtain better ttfb.

    Thanked by (1)skhron

    I believe in the good luck. Harder than I work luckier i get.

  • ChievoChievo OG

    @skhron said:

    @Chievo said:

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    In theory with paid plans you have access to more POPs and better speed . In theory at least from what i have read in different websites not sure at 100%

    Yeah but you don't need 100% of Cloudflare PoPs to lower your TTFB for Australia from 980ms to 440ms at least (already x2 faster!)

    @skhron said:

    @Chievo said:

    @skhron said:

    @Chievo said:

    @skhron said:
    @Chievo, probably I was wrong and issue is not in your configuration but due to high RTT. rocket.net is on cloudflare network and cached at the edge thus you can see such drastic difference in timings.

    Many thanks ! OK then next steps to be investigated how pro would improve it and try another cdns as well. to be investigated (soon).IDea and concept catched. Many thanks once again

    I am not sure you really need to buy Cloudflare Pro plan for caching. However, you should research how Cloudflare caching works and how to convince it to cache, for example, your homepage for unauthenticated visitors.

    In theory with paid plans you have access to more POPs and better speed . In theory at least from what i have read in different websites not sure at 100%

    Yeah but you don't need 100% of Cloudflare PoPs to lower your TTFB for Australia from 980ms to 440ms at least (already x2 faster!)

    IN fact checking their website from 5 USD i could get Argo Smart Routing and for another 5 usd another optimization pack for css ,permacache etc. Let s try from them

    I believe in the good luck. Harder than I work luckier i get.

  • bikegremlinbikegremlin ModeratorOGContent Writer

    I think that's normal if you aren't using a CDN that coppies everything (images and (cached) HTML).

    My Cloudflare Pro (and WordPress APO) experiment

    Thanked by (1)Chievo

    Relja of House Novović, the First of His Name, King of the Plains, the Breaker of Chains, WirMach Wolves pack member
    BikeGremlin's web-hosting reviews

  • ChievoChievo OG
    edited February 17

    @bikegremlin said:
    I think that's normal if you aren't using a CDN that coppies everything (images and (cached) HTML).

    My Cloudflare Pro (and WordPress APO) experiment

    yeah i would buy next week the argo of cloudflare+the optimizator from them+bunny cdn for images and leet s see the combo of them how would be . Trying and making mistakes,etc the way i have learnt the hard way :D

    Thanked by (1)bikegremlin

    I believe in the good luck. Harder than I work luckier i get.

  • FatGrizzlyFatGrizzly Hosting Provider

    @eva2000 can help you out, he/she is really talented at optimizing webservers

    Thanked by (1)Chievo
  • ChievoChievo OG

    @FatGrizzly said:
    @eva2000 can help you out, he/she is really talented at optimizing webservers

    Unfortunately not member here grrrr i have found. His/her contribution on cloudflare tldr pay more on cloudflare for more pops + another optimizations over cloudflare . Bunnycdn should move there and cause them some movement . it seems that cf is the giant right now .

    I believe in the good luck. Harder than I work luckier i get.

  • ChievoChievo OG
    edited February 17

    Tldr what eva said in centminmod:

    To fully optimise you need to optimize 3 segments.
    segment 1 - connection between visitor and CF edge server i.e. CDN cache/CDN cache control, WAF, Firewall, Page Rules, Mirage, Polish webP, HTTP/2, HTTP/3, CF Workers (i.e. custom/advanced caching) etc. Cloudflare CDN level caching effectively reduces your origin server resource usage loads for CPU/Memory etc as request load is offloaded to Cloudflare CDN edge servers in an Anycast manner to the closest CF datacenter to the visitor.
    segment 2 - connection between CF edge server and your origin i.e. TLSv1.3 origin server support, Argo, Railgun, Full SSL/ECDSA SSL certificates origin served, pre-compressed asset served from origin, Tiered Caching.
    segment 3 - your origin server’s performance/optimisations i.e. web server, PHP, MySQL server optimisations and server hardware specs. Origin server side caching, PHP Zend Opcache caching, MySQL’s various buffer caching and various other origin server side caching can also be used.
    Cloudflare can only help for segments 1 & 2 for cached guest/non-logged based visitors will easily scale. Now for Cloudflare CDN cache miss/bypass and logged in user for web apps like forums/wordpress, performance will be determined by segment 3. Which is the default with Cloudflare as dynamically generated HTML pages aren’t cached by default so cache miss means, whatever performance you have is measuring your origin server’s response time.

    Part2

    TTFB factors
    Geographically relative source and target distance = how close test/visitor is to your server
    DNS, server connection & response time
    MySQL or database backend performance/settings
    Server hardware

    I believe in the good luck. Harder than I work luckier i get.

  • ChievoChievo OG

    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    Thanked by (1)skhron

    I believe in the good luck. Harder than I work luckier i get.

  • bikegremlinbikegremlin ModeratorOGContent Writer

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    Thanked by (1)Chievo

    Relja of House Novović, the First of His Name, King of the Plains, the Breaker of Chains, WirMach Wolves pack member
    BikeGremlin's web-hosting reviews

  • skhronskhron Hosting Provider

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    well ttfb is repeated always at first place so it matters in my personal opinion.
    CLS is 0
    and LCP is 0,8 seconds average at worldwide.
    which website are you checking for TTI ? i would like to test it in different locations as well. WHich is your recomendation?
    thanks

    I believe in the good luck. Harder than I work luckier i get.

  • ChievoChievo OG

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    May be the optimizer for the cache at tieredge could make a difference. I have not checked it yet. I wish to make different setups after some time in order to make a diference.may be 7 or 15 days aprox to have some data

    I believe in the good luck. Harder than I work luckier i get.

  • skhronskhron Hosting Provider
    edited February 19

    @Chievo said:

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    May be the optimizer for the cache at tieredge could make a difference. I have not checked it yet. I wish to make different setups after some time in order to make a diference.may be 7 or 15 days aprox to have some data

    Cloudflare is not a magic black box that is able to know ahead of the time which JS code is not required for your application to remove it for optimization purposes :D

    Just doing classical minifying is not enough, as the visitors' resources are still spent to download extra scripts/styles, parse and execute them.

    Thanked by (1)Chievo

    Check our KVM VPS (flags are clickable): 🇵🇱 🇸🇪 | Looking glass: 🇵🇱 🇸🇪

  • ChievoChievo OG

    @skhron said:

    @Chievo said:

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    @skhron said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    I don't think it is something that can be significantly improved at a network level. AFAIK those values are about how optimized your CSS and JS.

    May be the optimizer for the cache at tieredge could make a difference. I have not checked it yet. I wish to make different setups after some time in order to make a diference.may be 7 or 15 days aprox to have some data

    Cloudflare is not a magic black box that is able to know ahead of the time which JS code is not required for your application to remove it for optimization purposes :D

    Just doing classical minifying is not enough, as the visitors' resources are still spent to download extra scripts/styles, parse and execute them.

    well i am happy with the routing .it comes at extra cost but i am happy for the moment . True well i am not sure what the next pack of optimization do at the back on cloudflre so i am trying and doing mistakes learning process. just checking improving or not :(

    I believe in the good luck. Harder than I work luckier i get.

  • bikegremlinbikegremlin ModeratorOGContent Writer

    @Chievo said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    well ttfb is repeated always at first place so it matters in my personal opinion.
    CLS is 0
    and LCP is 0,8 seconds average at worldwide.
    which website are you checking for TTI ? i would like to test it in different locations as well. WHich is your recomendation?
    thanks

    I use GTmetrix for testing, as it gives decent ballpark values that can be compared over time.

    Your page will most likely take about two seconds to load fully, and that is OK.
    You pay for Cloudflare to shave about 0.5 seconds at best (at the most remote location from your server).

    Thanked by (1)Chievo

    Relja of House Novović, the First of His Name, King of the Plains, the Breaker of Chains, WirMach Wolves pack member
    BikeGremlin's web-hosting reviews

  • ChievoChievo OG

    @bikegremlin said:

    @Chievo said:

    @bikegremlin said:

    @Chievo said:
    @bikegremlin and @skhron UPDATED:
    AFter using argo smarting plan for 5usd+0,10per gb

    Before Europe:120 .Now 102ms
    Before USA: 480 NOW 260ms
    Before Asia: 620ms Now 290ms

    Before Australia:910 Now 300ms

    Before Average 450ms Now 219ms

    One single click update on the cloudflare panel .Massive improvement over 50%on the average and 2 and x3 on asia and australia.

    More important factors (at least in my opinion) than the TTFB are:

    • CLS (cumilative shift layout) - anything under 0.1 is OK.
    • Largest contentful paint (LCP) - under 1 second is very good, 1.5s is OK.
    • Time to interactive (TTI) - under 2.5 seconds is good, under 1.5 is very very good.

    How do those fare now?
    I had some CLS problems with Cloudflare Pro.

    well ttfb is repeated always at first place so it matters in my personal opinion.
    CLS is 0
    and LCP is 0,8 seconds average at worldwide.
    which website are you checking for TTI ? i would like to test it in different locations as well. WHich is your recomendation?
    thanks

    I use GTmetrix for testing, as it gives decent ballpark values that can be compared over time.

    Your page will most likely take about two seconds to load fully, and that is OK.
    You pay for Cloudflare to shave about 0.5 seconds at best (at the most remote location from your server).

    Yes gtmetrix is/was nice. Now is fully paid option. Unfortunately :(

    Yes i am aware that this optimize the South Americans USA and australia. But i think that this is notable difference for them. My traffic comes mainly from Spain Latam USA UK and australia so it make sense . It depends from the countries from where the traffic comes

    I believe in the good luck. Harder than I work luckier i get.

  • AuroraZeroAuroraZero Moderator

    Depending on the site you can waterfall it to find your bottlenecks.

    Load said bottlenecks directly into your CDN and have them serve them faster.

    Only way I know of to shave anytime load time as the CDN "SHOULD" load those from the nearest source.

    Thanked by (2)Chievo bikegremlin

    Free Hosting at YetiNode | Cryptid Security | URL Shortener | LaunchVPS | ExtraVM | Host-C | In the Node, or Out of the Loop?

  • poppzpoppz

    @Chievo said: on the another hand rocket.net have 57ms for europe ttfb. 53ms for usa and 185 for asia and 43 for australia.

    >

    FWIW, Rocket.net use Cloudflare Enterprise for all of their sites, which by itself is a big help in reducing TTFB and in general and connections to the sites.

    Thanked by (1)Chievo
  • febryanvaldofebryanvaldo
    edited February 22

    What tools do you use to benchmark the TTFB? GTMetrix?

    Thanked by (1)Chievo

    KangServer - VPS Benchmark
    Android | Windows 11 | Ubuntu 22.04 | XShell

  • ChievoChievo OG

    @febryanvaldo said:
    What tools do you use to benchmark the TTFB? GTMetrix?

    Speedvitals and keycdn among others

    I believe in the good luck. Harder than I work luckier i get.

  • ChievoChievo OG

    @poppz said:

    @Chievo said: on the another hand rocket.net have 57ms for europe ttfb. 53ms for usa and 185 for asia and 43 for australia.

    >

    FWIW, Rocket.net use Cloudflare Enterprise for all of their sites, which by itself is a big help in reducing TTFB and in general and connections to the sites.

    Thanks for the clarification . yes they are unbeatable in that aspect :( deep pockets

    I believe in the good luck. Harder than I work luckier i get.

Sign In or Register to comment.

© LowEndSpirit 2024

ipv6 ready