1 hour ago, NilSS said:

Google actually have racks in the CAT IDC in Bangkok, but in any case, it's the peering and transit arrangements that matter, and for the same reasons I colo my own rack in Singapore.

 

Thailand has always had very good/fast data cable connections with Singapore, AFAIK.

 

In my experience, if your end point can get a good speed connection to Singapore, then the Thailand to Singapore hop should rarely be a problem.

 

35 minutes ago, Chelseafan said:

There is a difference between advertised speeds and actual speeds. I may be wrong but I doubt many get those sort of speeds consistantly but as I just stated in a previous post, upload speeds need to be improved. Can you run a speedtest ? Do you consistantly get 150 download speed ?

 

 

People with good gigabit fiber connections in Thailand can definitely get close to 1G download speeds for endpoints INSIDE Thailand, assuming their router, network card, ethernet cables, etc are all Gigabit capable. Not going to happen via most wifi, though.

 

But where you won't get Gigabit speeds or anywhere close to them is having a Gigabit ISP fiber service here and then trying to use that to connect to some international location like in the U.S. or Europe. In those cases, the available speeds from Thailand even with the best equipment come down considerably.

 

Domestic is one thing. International (except to Singapore) is entirely different.

 

1 hour ago, TallGuyJohninBKK said:

 

People with good gigabit fiber connections in Thailand can definitely get close to 1G download speeds for endpoints INSIDE Thailand, assuming their router, network card, ethernet cables, etc are all Gigabit capable. Not going to happen via most wifi, though.

 

But where you won't get Gigabit speeds or anywhere close to them is having a Gigabit ISP fiber service here and then trying to use that to connect to some international location like in the U.S. or Europe. In those cases, the available speeds from Thailand even with the best equipment come down considerably.

 

Domestic is one thing. International (except to Singapore) is entirely different.

 

Is that across Thailand or just in Bangkok ?

Romania?

10 hours ago, Chelseafan said:

Thinking about the article. Surely that's only based on people who do a test on Speednet ? What if only 1 person in thailand did that and he got a speed of 149.95 ?

 

And what if only one raindrop fell during a rain storm?   

 

Speedtest.net is the by far the most  well known and used speed tester....used by many, many, many people everyday.  Not saying it's the best tester, but surely the most used.  Heck, I bet it's used tens of thousands of times everyday just in Thailand.  I probably use it at least once a day just to confirm my internet connection is running full speed.

And take a look at below Nperf.com speed tester report for 2019 where over 10 million nperf.com speed tests were run against 3BB, AIS, TOT, and True.   The report talks the "spectacular increase of local fixed Internet speeds in only 1 year (2019)"   See the weblink below for the detailed/full report.

 

Heck, I'm always seeing Thai ISPs stringing more fiber optic lines along major roads and little sois. Seems over the last few years or so, there was literally an explosion of stringing fiber optics lines almost everywhere in Thailand.  But yea, I know, some folks who live way out in the sticks or older high rises still can't get fiber optics to their location....still limited to DSL of around 75Mb tops.

 

When I think back about a dozen years ago TOT 2Mb ADSL speed was the only/highest internet speed plan available to my western Bangkok moobaan; but starting around 2016 True came to my moobaan with Cable/DOCSIS with speeds up to 1Gb (very expensive then)....then came AIS Fibre with fiber optics....then came TOT fiber optics....then came 3BB fiber optics...now True is replacing Cable/DOCSIS in my moobaan with fiber optics.  Thailand 1Gb plans are now pretty much dirt cheap....much cheaper than most western countries.  

 

https://media.nperf.com/files/publications/TH/2020-01-15_fixed-internet-connections-survey-nPerf-2019.pdf

 

19 minutes ago, Pib said:

And take a look at below Nperf.com speed tester report for 2019 where over 10 million nperf.com speed tests were run against 3BB

 

I know I have mentioned this in other threads. Mid 2018 I had serious issues with my 3BB 30/10 VDSL, which they claimed was because my house is located 1.4 Km from the Dslam, which indeed stretches it a bit.

Nevertheless, on a particular day I had the full 3bb technical team for a visit to inspect the issues.

 

The team existed out of the top technicians from 3 regions, so people who are supposed to know what they are doing.

 

At one point they managed to get 150 Mbps download speeds, using the Nperf speed test, while on their own website they would not even get 30 Mbps. They realized that result multiple times, so it was definitely not a 1 time error

 

Other than that my package only was 30 Mbps, and that I'm in a poor infrastructure country at the maximum distance of the Dslam, even under ideal circumstances a copper wire, as used for ADSL / VDSL transmissions, maximum capacity is around 80 - 90 Mbps.

 

So far for the credibility of Nperf.

 

 

 

 

 

27 minutes ago, Susco said:

 

I know I have mentioned this in other threads. Mid 2018 I had serious issues with my 3BB 30/10 VDSL, which they claimed was because my house is located 1.4 Km from the Dslam, which indeed stretches it a bit.

Nevertheless, on a particular day I had the full 3bb technical team for a visit to inspect the issues.

 

The team existed out of the top technicians from 3 regions, so people who are supposed to know what they are doing.

 

At one point they managed to get 150 Mbps download speeds, using the Nperf speed test, while on their own website they would not even get 30 Mbps. They realized that result multiple times, so it was definitely not a 1 time error

 

Other than that my package only was 30 Mbps, and that I'm in a poor infrastructure country at the maximum distance of the Dslam, even under ideal circumstances a copper wire, as used for ADSL / VDSL transmissions, maximum capacity is around 80 - 90 Mbps.

 

So far for the credibility of Nperf.

 

 

 

 

 

Back in 2018 the 3BB speed tester was still flash-based.   Nperf.com has always been HTML5 based and the 3BB speed tester is now HTML5 based.  I expect the differences you saw were due to one tester being flashed based and the other HTML5 based combined with the problems associated with your connection.  Flash based testers are less accurate and subject to a variety of other problems like security vulnerabilities. 

Most people use provided router on 2.4ghz wifi radio. Otherwise, Thailand would surpass other countries.

 

I get 1gbit within Thailand and 650 Mbps to Singapore using speedtest-cli, which is pretty darn good.

 

I've been in Thailand since 2011 and the improvements are amazing. So much changed since 2011 and Thailand is pretty close to Singapore internet speeds. Singapore is luckier because it is a densely populated city state (and public housing!) and it hosts so many peering points and datacenter from large companies. 

 

Considering the cost of electricity and labor, Thailand should take over in this regard but it is highly unlikely due to laws and regulations and computer act.

15 minutes ago, Pib said:

Back in 2018 the 3BB speed tester was still flash-based.   Nperf.com has always been HTML5 based and the 3BB speed tester is now HTML5 based.  I expect the differences you saw were due to one tester being flashed based and the other HTML5 based combined with the problems associated with your connection.  Flash based testers are less accurate and subject to a variety of other problems like security vulnerabilities. 

Going by your own explanation, the Nperf should have been the more reliable, yet it was Nperf that gave the 150 Mbps download speed.

 

Can you provide any credible resource that shows that a copper wire, of the type used for ADSL/VDSL is capable of a 150 Mbps data transfer over any length of wire?

 

Other than that my package was 30Mbps, and the maximum package 3BB had at time was 50Mbps, but they wouldn't sign me up for that one because they knew the distance between Dslam and my house was too far.

root@Synology:~# ./speedtest --server-id=367

   Speedtest by Ookla

     Server: NewMedia Express - Singapore (id = 367)
        ISP: 3BB Broadband
    Latency:    36.09 ms   (1.21 ms jitter)
   Download:   668.26 Mbps (data used: 1.1 GB)                               
     Upload:   904.78 Mbps (data used: 1.1 GB)                               
Packet Loss:     0.0%
 Result URL: https://www.speedtest.net/result/c/ae1def8a-82f6-4007-8935-4e3db44754d2

I also think with Thailand being ranked #4 in speed that's it's truly a big achievement since the speed represents an average sized country with almost a 70M population.   

 

Although Singapore and Hong Kong are rated #1 and #2, respectively, considering they are both just city-states/small land masses and less than 8M population for each.....stringing a bunch of fiber optics across their city size land mass was pretty easy for them compared to stringing fiber optics throughout a larger country with a much higher population.  Typically much easier for a small, low population country to achieve higher country-wide internet speeds than a large country....large in population and/or land mass.

 

Heck, throw out Singapore and Hong Kong from the rankings and Thailand is #2.

13 hours ago, FritsSikkink said:

2 seconds

2.x seconds

3BB, 1000/500.

 

I was also surprised when I selected a German based "official" speedtest site which is meant to profile the ISPs.

To get through I select some arbitrary configuration (the results will likely be ignored with a Thai IP).

 

In this case I pretend to have a fairly expensive "business tariff" 200/40 Mbit/s and easily surpassed these values with 294/88 Mbit/s, ping 179 ms.

 

Really a lot has changed. When I first started with internet at home here in 2011 I had to use GPRS (if some remember what this is )

10 minutes ago, Susco said:

Going by your own explanation, the Nperf should have been the more reliable, yet it was Nperf that gave the 150 Mbps download speed.

 

Can you provide any credible resource that shows that a copper wire, of the type used for ADSL/VDSL is capable of a 150 Mbps data transfer over any length of wire?

 

Other than that my package was 30Mbps, and the maximum package 3BB had at time was 50Mbps, but they wouldn't sign me up for that one because they knew the distance between Dslam and my house was too far.

Consistent max DSL speed is around 100Mb over a short and high quality copper line.  But for short periods like say a 15 second speed test I expect 150Mb could be reached for short bursts.   Additionally, the Nperf.com aloghtrim reports both max/highest speed (like bursts) and average speed whereas speedtest.net/OOKLA based testers report more of an average.  

Thank you Huawei for providing the technology to Thailand to make this possible.

2 minutes ago, mistral53 said:

Thank you Huawei for providing the technology to Thailand to make this possible.

Huawei did not invent Fiber or ONT/GPON. They only sell it cheap so they can control it. 

root@Synology:~# ./speedtest --server-id=4235    

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: 3BB Broadband
    Latency:    36.23 ms   (0.94 ms jitter)
   Download:   773.14 Mbps (data used: 1.1 GB)                               
     Upload:   882.23 Mbps (data used: 1.4 GB)                               
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/135f2006-cce0-46d9-ade8-bfabc204be6b

 

Wanted to verify this wasn't a one time thing so I did a retest

 

root@Synology:~# ./speedtest --server-id=4235

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: 3BB Broadband
    Latency:    35.50 ms   (0.26 ms jitter)
   Download:   801.38 Mbps (data used: 1.3 GB)                               
     Upload:   901.87 Mbps (data used: 1.6 GB)                               
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/96ff8dc0-d08f-442b-9524-5ccefc0bc1dc

 

Oh yeah.

10 minutes ago, Pib said:

Consistent max DSL speed is around 100Mb over a short and high quality copper line.

 

Surely doesn't apply then to a user 1.4km from Dslam and if you see the cable installation here that means probably 1.8km of wire, and in a country where the lowest quality of materials is used.

 

13 minutes ago, Pib said:

But for short periods like say a 15 second speed test I expect 150Mb could be reached for short bursts.

 

So no credible resource then? And apart from that 3BB has never provided speeds in that range over VDSL

 

https://computer.howstuffworks.com/vdsl2.htm

 

VDSL operates over the copper wires in your phone line in much the same way that ADSL does, but there are a couple of distinctions. VDSL can achieve incredible speeds, as high as 52 Mbps downstream (to your home) and 16 Mbps upstream (from your home). That is much faster than ADSL, which provides up to 8 Mbps downstream and 800 Kbps (kilobits per second) upstream. But VDSL's amazing performance comes at a price: It can only operate over the copper line for a short distance, about 4,000 feet (1,200 m).

 

 

No understand why Thailand is not

the N. 1 ... this time ????

7 minutes ago, Susco said:

 

Surely doesn't apply then to a user 1.4km from Dslam and if you see the cable installation here that means probably 1.8km of wire, and in a country where the lowest quality of materials is used.

 

 

So no credible resource then? And apart from that 3BB has never provided speeds in that range over VDSL

 

https://computer.howstuffworks.com/vdsl2.htm

 

VDSL operates over the copper wires in your phone line in much the same way that ADSL does, but there are a couple of distinctions. VDSL can achieve incredible speeds, as high as 52 Mbps downstream (to your home) and 16 Mbps upstream (from your home). That is much faster than ADSL, which provides up to 8 Mbps downstream and 800 Kbps (kilobits per second) upstream. But VDSL's amazing performance comes at a price: It can only operate over the copper line for a short distance, about 4,000 feet (1,200 m).

 

 

That article is not even dated (or I didn't see a date) and surely "old"....it was probably talking the original VDSL standard; not VDSL2 that ISPs have used for many years.   See below Wikipedia article regarding max speeds for the VDSL standards.    

 

https://en.wikipedia.org/wiki/VDSL

 

 

Puts most western countries to shame. I get 1 gigabyte downloads for under $20 a month. Well done Thailand.

 

9 minutes ago, Pib said:

That article is not even dated (or I didn't see a date) and surely "old"....it was probably talking the original VDSL standard; not VDSL2 that ISPs have used for many years.   See below Wikipedia article regarding max speeds for the VDSL standards.    

 

https://en.wikipedia.org/wiki/VDSL

 

 

 

Yes most likely the article is old, as is VDSL, but 3BB has never used VDSL 2, and since I still have the router here, I can see it is not VDSL2 capable.

 

3BB currently still use VDSL here in Pattaya, and the highest package they offer is 50/20. Why would that be if they are capable of 200 Mbps?

 

True also offers VDSL, with the biggest package 50 Mbps

@Pib  and this is an article updated 5 days ago.

 

https://help.sonic.com/hc/en-us/articles/115007031987-What-Speeds-Can-I-Expect-on-ADSL2-VDSL2

 

VDSL2

The maximum throughput (speed) for VDSL2 is 100Mbs download and 100Mb/s upload, though it is rare to achieve these figures outside of lab conditions. VDSL2 yields better speeds than ADSL2+ at distances shorter than 3,000 feet. The method of data transmission for VDSL2 is PTM (packet transfer mode) with a bandwidth overhead of approximately 5%.

My 4g here averages 40-65 mbps. In the US, at more than double the cost, Verizon averages 10-15 mbps. My True fiber at home averages 540 down, 580 up! At a third the cost of service in the US. Thailand had done a remarkable job with cellular and fiber. Amazing. 

 

https://www.cnet.com/how-to/how-to-find-a-reliable-network-speed-test/

Does the type of speed test make a difference?

Many experts claim HTML5-based speed tests are more accurate than tests that use Java and Adobe Flash. Others point out that multithread tests such as those used by Ookla (Speedtest.net and branded by many ISPs) don't represent real-world network traffic as well as single-thread tests.

31 minutes ago, Susco said:

 

 

Yes most likely the article is old, as is VDSL, but 3BB has never used VDSL 2, and since I still have the router here, I can see it is not VDSL2 capable.

 

3BB currently still use VDSL here in Pattaya, and the highest package they offer is 50/20. Why would that be if they are capable of 200 Mbps?

 

True also offers VDSL, with the biggest package 50 Mbps

 

Sure 3BB offered VDSL2 and VDSL2 modems....they just didn't' advertise it as such because it was just too technical of language/advertising for most folks.  Usually it was just advertised as DSL....just didn't get technical with VDSL2, VDSL2+, etc., terminology.  See below modem-router manual/image from the 3BB support webpage for their HG630 VDSL2 modem-router. 

 

Thai ISPs have been using VDSL2 for long time but due to the generally <deleted> poor quality and purely maintained Thai copper phone/xDSL lines/DSLAMs/junction boxes they generally only sold xDSL speed plans up to 50Mb....but some did sell from 75Mb packages depending on the location of your residence like if you lived in a large city/highrise.   

 

 

http://www.3bb.co.th/uploads/20180411093041HG630V2.pdf

-thailand-has-4th-fastest-internet-speeds-in-the-world

 

Not on my computer. waiting for ads to download all the time.

25 minutes ago, Pib said:

 

Sure 3BB offered VDSL2 and VDSL2 modems....they just didn't' advertise it as such because it was just too technical of language/advertising for most folks.  Usually it was just advertised as DSL....just didn't get technical with VDSL2, VDSL2+, etc., terminology.  See below modem-router manual/image from the 3BB support webpage for their HG630 VDSL2 modem-router. 

 

Thai ISPs have been using VDSL2 for long time but due to the generally <deleted> poor quality and purely maintained Thai copper phone/xDSL lines/DSLAMs/junction boxes they generally only sold xDSL speed plans up to 50Mb....but some did sell from 75Mb packages depending on the location of your residence like if you lived in a large city/highrise.   

 

 

http://www.3bb.co.th/uploads/20180411093041HG630V2.pdf

OK, let's agree to disagree.

 

Still doesn't explain why Nperf was able to show a 150 Mbps download on my connection which obviously has the worst of worst conditions possible.

 

Simple answer, because the speed test is fake.

 

I'm sure you are also aware that ISP can manipulate the test results.

 

https://www.ispreview.co.uk/index.php/2016/01/ookla-accuses-uk-broadband-isp-manipulating-internet-speedtests.html

Need better routers to take advantage of it though 

2 hours ago, muratremix said:

root@Synology:~# ./speedtest --server-id=4235    

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: 3BB Broadband
    Latency:    36.23 ms   (0.94 ms jitter)
   Download:   773.14 Mbps (data used: 1.1 GB)                               
     Upload:   882.23 Mbps (data used: 1.4 GB)                               
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/135f2006-cce0-46d9-ade8-bfabc204be6b

 

Wanted to verify this wasn't a one time thing so I did a retest

 

root@Synology:~# ./speedtest --server-id=4235

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: 3BB Broadband
    Latency:    35.50 ms   (0.26 ms jitter)
   Download:   801.38 Mbps (data used: 1.3 GB)                               
     Upload:   901.87 Mbps (data used: 1.6 GB)                               
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/96ff8dc0-d08f-442b-9524-5ccefc0bc1dc

 

Oh yeah.

 

 

Well, I can't match the download "and" upload speed of your 3BB 1Gb/1Gb speed plan to Singapore but with my AIS 500/200 plan speed toggled to 700/50 I'm still doing pretty good....can match your download speeds plus or minus a hair.   Tested twice to the same server as you...using speedtest cli.

 

 

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: AIS Fibre
    Latency:    34.55 ms   (0.41 ms jitter)
   Download:   778.58 Mbps (data used: 1.0 GB)
     Upload:    58.26 Mbps (data used: 42.0 MB)
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/4f0ef36e-a638-4eb2-b670-40b56ba1bd17

 

C:\Users\Downloads\Speedtest CLI>speedtest -s 4235

   Speedtest by Ookla

     Server: StarHub Mobile Pte Ltd - Singapore (id = 4235)
        ISP: AIS Fibre
    Latency:    33.58 ms   (0.68 ms jitter)
   Download:   793.78 Mbps (data used: 988.3 MB)
     Upload:    56.91 Mbps (data used: 54.2 MB)
Packet Loss: Not available.
 Result URL: https://www.speedtest.net/result/c/7307260c-1ba6-4f4b-b3a0-4b31daf3aa48