The CryptoStorm Speed Test Thread
The CryptoStorm Speed Test Thread
There have been a few users who have documented that they are being capped at ~37.6Mb/s (4.7MB/s) on a 100Mb/s (12.5MB/s) connection, regardless of the node they connect to. In order to build a picture of the performance available to network members, I've created this thread so we can collate experiences.
Guidelines/Rules
We are trying to build an accurate picture of real world throughput, so here are some rules.
1. (Obvious) Minimise the volume of other network traffic on your connection for the duration of the test.
2. DO NOT use speedtest.net. Since we're trying to measure real world throughput, please use a real world application, such as a web browser doing an HTTP download, BitTorrent client, FTP client (FileZilla) or even curl or wget.
3. Where possible, try to use the node which is geographically closest to you. Technically, with fast enough (and few enough) inter-links this should make no difference, but reality is a different story.
4. Again where possible, use the source file (see below) which is geographically closest to you.
Source Files
Linux Mint 17.1 Cinnamon (.torrent is here also)
LinHost Test Files
Guidelines/Rules
We are trying to build an accurate picture of real world throughput, so here are some rules.
1. (Obvious) Minimise the volume of other network traffic on your connection for the duration of the test.
2. DO NOT use speedtest.net. Since we're trying to measure real world throughput, please use a real world application, such as a web browser doing an HTTP download, BitTorrent client, FTP client (FileZilla) or even curl or wget.
3. Where possible, try to use the node which is geographically closest to you. Technically, with fast enough (and few enough) inter-links this should make no difference, but reality is a different story.
4. Again where possible, use the source file (see below) which is geographically closest to you.
Source Files
Linux Mint 17.1 Cinnamon (.torrent is here also)
LinHost Test Files
Re: The CryptoStorm Speed Test Thread
@thread
OK, so I'll go first.
Connection: 8Mb/1Mb ADSL
Exit Node: Turing (London)
(SoftLayer, Amsterdam)
OK, so I'll go first.
Connection: 8Mb/1Mb ADSL
Exit Node: Turing (London)
Code: Select all
wget -O /dev/null http://speedtest.ams01.softlayer.com/downloads/test100.zip
2015-06-08 12:00:38 (831 KB/s) - ‘/dev/null’ saved [104874307/104874307]
- marzametal
- Posts: 434
- Joined: Mon Aug 05, 2013 11:39 am
Re: The CryptoStorm Speed Test Thread
How connection is made:
- ADSL2+ ~24Mb/1Mb <--- too far from exchange so it's like ADSL actually...
- Singapore exit node, Singapore download source (apart from torrent... not sure where that originates from...)
- Windows Widget v2.22... UDP 443
- download via Firefox 29 between 100kb/s and 120kb/s
- download via Internet Download Manager between 650kb/s and 850kb/s
- download via Tixati between 850kb/s and 1000kb/s
- browser download is higher on other exit nodes, but only reaches 200kb/s - 250kb/s <--- ISP throttling?
- download manager app can hit speeds of 1000kb/s to 1100kb/s and stay at that speed, if I connect to other exit nodes
- torrent client can max out and stay at 1100kb/s on other exit nodes
- I tried connecting via random UDP port (65501); the only difference I noticed was crisper and quicker webpage retrieval and response/reload
Re: The CryptoStorm Speed Test Thread
Frankfurt seems to have difficulties connecting to peers when using public torrents, private tracker no problem other than speed being quite modest 6-7MB/s.
With Frankfurt Linhost test file 1000MB from netherlands worked quite well at 8.5-9.0MB/s
And the same file with London is steadily at 3.9MB/s for the whole download, the torrent never went past 70KB/s.
Without CS my torrent speeds were 9MB/s and the linhost test file 10MB/s.
I still remain hopeful that there's some solution to this, or at least a good explanation.
With Frankfurt Linhost test file 1000MB from netherlands worked quite well at 8.5-9.0MB/s
And the same file with London is steadily at 3.9MB/s for the whole download, the torrent never went past 70KB/s.
Without CS my torrent speeds were 9MB/s and the linhost test file 10MB/s.
I still remain hopeful that there's some solution to this, or at least a good explanation.
Re: The CryptoStorm Speed Test Thread
@turbz
Can you clarify that the speeds you quote are MB/s as opposed to Mb/s? It would be good to keep things consistent for ease of readability. Thanks.
EDIT:
Connection: 8Mb/1Mb ADSL
Exit Node: Cantus (Frankfurt)
KTorrent 4.3.1 downloading the Linux Mint 17.1 torrent averaged 790KB/s (6.3Mb/s)
Can you clarify that the speeds you quote are MB/s as opposed to Mb/s? It would be good to keep things consistent for ease of readability. Thanks.

EDIT:
Connection: 8Mb/1Mb ADSL
Exit Node: Cantus (Frankfurt)
KTorrent 4.3.1 downloading the Linux Mint 17.1 torrent averaged 790KB/s (6.3Mb/s)
Re: The CryptoStorm Speed Test Thread
They are MB/s, at least thats how chrome says.
Re: The CryptoStorm Speed Test Thread
@turbz
Thanks for that. Also, something else came to mind: some of the exit nodes have 100Mb/s network interfaces, rather than 1Gb/s interfaces. The cap you're seeing on the London node may be a result of that (or perhaps something else). I'll have to poke staff to see which nodes have which interfaces.
Thanks for that. Also, something else came to mind: some of the exit nodes have 100Mb/s network interfaces, rather than 1Gb/s interfaces. The cap you're seeing on the London node may be a result of that (or perhaps something else). I'll have to poke staff to see which nodes have which interfaces.
Re: The CryptoStorm Speed Test Thread
Thanks a lot for the quick answers and the effort 

Re: The CryptoStorm Speed Test Thread
Ran a test today from the Kansas City node to linode test file in Dallas.
¡Qué miserables!
Connection: ADSL 22Mb/2Mb
¡Qué miserables!
Connection: ADSL 22Mb/2Mb
Code: Select all
curl -o /dev/null http://speedtest.dal01.softlayer.com/downloads/test100.zip
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 100M 100 100M 0 0 47070 0 0:37:08 0:37:08 --:--:-- 67877
Re: The CryptoStorm Speed Test Thread
@sysfu
Could you edit your post to include the supported bandwidth (8Mb/s, 24Mb/s, 100Mb/s) of your Internet connection? Many thanks.
Could you edit your post to include the supported bandwidth (8Mb/s, 24Mb/s, 100Mb/s) of your Internet connection? Many thanks.

Re: The CryptoStorm Speed Test Thread
Code: Select all
curl -o /dev/null http://mirror.nl.leaseweb.net/speedtest/100mb.bin
Connection: 200Mb/20Mb Cable
without VPN
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 95.3M 100 95.3M 0 0 21.2M 0 0:00:04 0:00:04 --:--:-- 21.2M
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 95.3M 100 95.3M 0 0 7347k 0 0:00:13 0:00:13 --:--:-- 6977k
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 95.3M 100 95.3M 0 0 487k 0 0:03:20 0:03:20 --:--:-- 582k
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 95.3M 100 95.3M 0 0 5026k 0 0:00:19 0:00:19 --:--:-- 3873k
Re: The CryptoStorm Speed Test Thread
Forgive me if I'm being dense but I no longer see an option to edit the previous post even though I am logged into the forum.parityboy wrote:@sysfu
Could you edit your post to include the supported bandwidth (8Mb/s, 24Mb/s, 100Mb/s) of your Internet connection? Many thanks.
The 'raw' speed of the Internet connection is ~22Mps down and ~2Mps up. Currently running running OpenVPN + OpenBSD 5.7 on an Alix device which serves entire home LAN.
I just tried another quick and dirty download test from a Windows workstation (with a youtube vid playing mind you) and the results were much better.
Code: Select all
aria2c http://speedtest.dal01.softlayer.com/downloads/test100.zip
Download Results:
gid |stat|avg speed |path/URI
======+====+===========+=======================================================
4735ca|OK | 341KiB/s|C:/Users/test/Downloads/test100.zip
I'll also try tearing down the Cryptostorm OpenVPN tunnel on the router and connect directly from both Windows and Mac systems using the current Narwhal widget and Viscosity respectively, to see what kind of results I get.
Last edited by sysfu on Wed Jun 10, 2015 10:08 pm, edited 3 times in total.
Re: The CryptoStorm Speed Test Thread
@sysfu
Ahh, yes I forgot the edit option becomes unavailable after a certain amount of time. No problem.
Ahh, yes I forgot the edit option becomes unavailable after a certain amount of time. No problem.

-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
Code: Select all
curl -o /dev/null http://mirror.nl.leaseweb.net/speedtest/100mb.bin
Code: Select all
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 95.3M 100 95.3M 0 0 6208k 0 0:00:15 0:00:15 --:--:-- 7315k
- Attachments
-
- torrent.jpg (16.84 KiB) Viewed 393043 times
-
- st-peter-speedtest-yt.JPG (25.93 KiB) Viewed 393043 times
Re: The CryptoStorm Speed Test Thread
Direct via ISP
OpenBSD/Alix router via Seattle cluster:
Narwhal 2.22 widget via Seattle cluster:
Code: Select all
aria2c http://speedtest.sea01.softlayer.com/downloads/test100.zip
Download Results:
gid |stat|avg speed |path/URI
======+====+===========+=======================================================
ed2189|OK | 2.6MiB/s|D:/Users/test/Downloads/test100.zip
Code: Select all
Download Results:
gid |stat|avg speed |path/URI
======+====+===========+=======================================================
a40207|OK | 415KiB/s|D:/Users/test/Downloads/test100.zip
Code: Select all
Download Results:
gid |stat|avg speed |path/URI
======+====+===========+=======================================================
463ae2|OK | 1.1MiB/s|D:/Users/test/Downloads/test100.zip
Re: The CryptoStorm Speed Test Thread
What's your "naked" speed?thisis4thepeople wrote:Exit node: LaikaCode: Select all
curl -o /dev/null http://mirror.nl.leaseweb.net/speedtest/100mb.bin
<snip>
Re: The CryptoStorm Speed Test Thread
Raw or nekkid connection speed: ADSL ~22Mbps/2Mbps
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
100/100 Mbps.parityboy wrote:What's your "naked" speed?thisis4thepeople wrote:Exit node: LaikaCode: Select all
curl -o /dev/null http://mirror.nl.leaseweb.net/speedtest/100mb.bin
<snip>
Re: The CryptoStorm Speed Test Thread
Thought I'd share a pretty cool command line client for speedtest.net that I found the other day: https://github.com/zpeters/speedtest
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
Another test: One only on YouTube, the other two simultaneous downloads (YouTube + savefrom.net).
67 Mbps on YouTube = 8,375 MB/s
6,8 MB + 16 Mbps = 8,8 MB/s
67 Mbps on YouTube = 8,375 MB/s
6,8 MB + 16 Mbps = 8,8 MB/s
Re: The CryptoStorm Speed Test Thread
I have a 152Mbps connection at home although it peaks short of 120Mbps IRL. While I'm connected via CStorm I see a top speed of around 50Mbps. The average speed for torrents is around 2.4MiB/s (unchanged from before my connection upgrade) and for regular files I've seen it peak at around 4MiB/s. I've been messaging through Twitter since it's seen as an edge case, difficult to tune my OS since Windows makes it hard to.
Re: The CryptoStorm Speed Test Thread
(Can't edit my posts since the forum upgrade so I'll make a new one.)
I've had to switch to a wired connection, with a 20ft Ethernet cable, because my wireless adapters either drop after 5 minutes or cause Windows to malfunction. On torrents I see ~5MiB/s for download and for regular files it's ~6MiB/s. The highest speed I can achieve on a naked connection is ~14MiB/s.
I've had to switch to a wired connection, with a 20ft Ethernet cable, because my wireless adapters either drop after 5 minutes or cause Windows to malfunction. On torrents I see ~5MiB/s for download and for regular files it's ~6MiB/s. The highest speed I can achieve on a naked connection is ~14MiB/s.
Re: The CryptoStorm Speed Test Thread
So I'm in my first few days of cryptostorm and it's pretty painful for me so far. Here's why.
Connection Details:
250 Mb down/20 Mb up - Eastern-ish Canada
All tests performed using wget of http://speedtest.dal01.softlayer.com/do ... est100.zip
No VPN
File wasn't completely large enough to show my normal sustained speed but the peak at the end is pretty typical and even a bit on the low side.
Montreal Exit - 198.27.76.1
So that was pretty awful. Also, I couldn't even keep a persistent SSH or IRC session without constant disconnects. So, let's try a completely different exit...
US Central - 142.54.172.51
Seriously, this was going between 5-20KB/sec if at all so I cancelled it.
NEXT. Let's see if going across the ocean helps...
Iceland - 109.71.42.163
Well that was substantially better and it finished strong at just over 4 MB/s. Let's try one more.
London - 130.180.201.117
That's more like it.
For the sake of comparison, let's check a couple exit nodes on PIA:
PIA - Toronto (Closest geographically)
PIA - US East
And one overseas for good measure, let's try London.
PIA - London
And that's that.
So, I'm not particularly thrilled with the North American exits that I tried, but at least Iceland and London performed alright.
Connection Details:
250 Mb down/20 Mb up - Eastern-ish Canada
All tests performed using wget of http://speedtest.dal01.softlayer.com/do ... est100.zip
No VPN
Code: Select all
100%[===============================================>] 104,874,307 21.4MB/s in 8.4s
2015-07-07 20:27:34 (11.9 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Montreal Exit - 198.27.76.1
Code: Select all
2015-07-07 20:53:34 (77.0 KB/s) - ‘/dev/null’ saved [104874307/104874307]
US Central - 142.54.172.51
Seriously, this was going between 5-20KB/sec if at all so I cancelled it.
NEXT. Let's see if going across the ocean helps...
Iceland - 109.71.42.163
Code: Select all
2015-07-07 21:03:35 (1.93 MB/s) - ‘/dev/null’ saved [104874307/104874307]
London - 130.180.201.117
Code: Select all
2015-07-07 21:01:30 (4.73 MB/s) - ‘/dev/null’ saved [104874307/104874307]
For the sake of comparison, let's check a couple exit nodes on PIA:
PIA - Toronto (Closest geographically)
Code: Select all
2015-07-07 21:05:43 (3.32 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Code: Select all
2015-07-07 21:06:54 (2.95 MB/s) - ‘/dev/null’ saved [104874307/104874307]
PIA - London
Code: Select all
2015-07-07 21:10:05 (1.74 MB/s) - ‘/dev/null’ saved [104874307/104874307]
So, I'm not particularly thrilled with the North American exits that I tried, but at least Iceland and London performed alright.
- marzametal
- Posts: 434
- Joined: Mon Aug 05, 2013 11:39 am
Re: The CryptoStorm Speed Test Thread
This one was a pain for me this morning... the others seem to serve me well.Thracky wrote: US Central - 142.54.172.51
Seriously, this was going between 5-20KB/sec if at all so I cancelled it.
Re: The CryptoStorm Speed Test Thread
@Thracky
On your faster speed runs via CS, what was your CPU load like?
On your faster speed runs via CS, what was your CPU load like?
Re: The CryptoStorm Speed Test Thread
One core bouncing around between 9-25% on the openvpn process. (i7-2600k, no overclock)parityboy wrote:@Thracky
On your faster speed runs via CS, what was your CPU load like?
Re: The CryptoStorm Speed Test Thread
Well, 25% load tells me the issue certainly isn't at your end. I'm beginning to wonder if the issue is the OpenVPN process itself, or maybe further down the networking stack? Considering the fact that Linux-based routers are capable of throughput far greater than 50Mbit/s - combined with the fact that CS' exit nodes are full-blown x86 servers - I'd say the throughput issue lies with OpenVPN itselfThracky wrote:One core bouncing around between 9-25% on the openvpn process. (i7-2600k, no overclock)parityboy wrote:@Thracky
On your faster speed runs via CS, what was your CPU load like?
However, without proper testing I'm just guessing.

Re: The CryptoStorm Speed Test Thread
Using exit linux-cantus1.xn--cdaan2d.be46.165.222.248
Native 22Mbps/2Mbps dn/up
Native 22Mbps/2Mbps dn/up
Code: Select all
Retrieving speedtest.net configuration...
Retrieving speedtest.net server list...
Testing from Leaseweb Germany GmbH (46.165.222.248)...
Selecting best server based on latency...
Hosted by Base-Mail (Frankfurt) [100.73 km]: 237.964 ms
Testing download speed........................................
Download: 8.62 Mbit/s
Testing upload speed..................................................
Upload: 0.98 Mbit/s
Re: The CryptoStorm Speed Test Thread
@sysfu
From the OP
From the OP
2. DO NOT use speedtest.net. Since we're trying to measure real world throughput, please use a real world application, such as a web browser doing an HTTP download, BitTorrent client, FTP client (FileZilla) or even curl or wget.
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread

64 Mbit/s on UK server on a 100 Mbit/s line.
- DonderPjem
- Posts: 5
- Joined: Wed Sep 23, 2015 6:14 am
Re: The CryptoStorm Speed Test Thread
My tradeoff for using CS is almost 90% of my bandwidth. My usually 100mbit up/down is reduced to 10~12mbit. Fortunately I only have to endure this during internet browsing as I outsource my heavy torrenting to a seedbox. But it still becomes noticable when trying to watch yt videos in 1440p or download something larger than a few megs.
It looks like I am being hit the hardest here, does anyone have optimalisation tips? Should I just hope for a server cluster in my country?
It looks like I am being hit the hardest here, does anyone have optimalisation tips? Should I just hope for a server cluster in my country?
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
Preferably, you'd choose a server close to you. Test all servers closest to you and use the one that gives you the highest speed, if that is what you prefer.
- DonderPjem
- Posts: 5
- Joined: Wed Sep 23, 2015 6:14 am
Re: The CryptoStorm Speed Test Thread
Im in the Netherlands and the london, frankfurt and paris clusters all give the same performance.
Re: The CryptoStorm Speed Test Thread
@DonderPjem
Can you give a little more details as to your setup?
1. Is the VPN endpoint on your router or on your desktop? What's the respective CPU load?
2. What CPU/OS are you running?
Thanks.
Can you give a little more details as to your setup?
1. Is the VPN endpoint on your router or on your desktop? What's the respective CPU load?
2. What CPU/OS are you running?
Thanks.

- DonderPjem
- Posts: 5
- Joined: Wed Sep 23, 2015 6:14 am
Re: The CryptoStorm Speed Test Thread
I'm using the cryptostorm widget on my 64bit windows machine with an i7 2630QM. The cpu usage is insignificant. I'm pretty sure the bottleneck is not on my part. PIA gave me speeds at around 95% of my native bandwidth. Ive tried every server cluster by now and I havent been able to utilize more than 15% of my 100mbit connection yet.
Re: The CryptoStorm Speed Test Thread
@DonderPjem
Thanks for the info.
Would it be possible for you to conduct a couple of tests using the links at the top of the page? Basically, connect to an exit node closest to your location, then pick a file source for Linux Mint which is also closest to your location - so for example if you are in Germany, you would use the Frankfurt exit node and download the ISO from this link.
If you could report back with the results of a couple of runs, it would be appreciated.
Thanks for the info.

Would it be possible for you to conduct a couple of tests using the links at the top of the page? Basically, connect to an exit node closest to your location, then pick a file source for Linux Mint which is also closest to your location - so for example if you are in Germany, you would use the Frankfurt exit node and download the ISO from this link.
If you could report back with the results of a couple of runs, it would be appreciated.

- DonderPjem
- Posts: 5
- Joined: Wed Sep 23, 2015 6:14 am
Re: The CryptoStorm Speed Test Thread
Thanks for your time!
Ive tried every mirror with the respective exit node and note down the maximum speed it maintained for about 10 seconds in the first 2 minutes.
(speeds are in KB/s and mirrors are in order of appearance on http://www.linuxmint.com/edition.php?id=172)
france
300
250
400
250
600
germany
300
900
200
350
300
700
600
uk
400
offline
switzerland
550
netherlands (native connection)
3000
2200
netherlands (germany exit node)
500
350
Ive tried every mirror with the respective exit node and note down the maximum speed it maintained for about 10 seconds in the first 2 minutes.
(speeds are in KB/s and mirrors are in order of appearance on http://www.linuxmint.com/edition.php?id=172)
france
300
250
400
250
600
germany
300
900
200
350
300
700
600
uk
400
offline
switzerland
550
netherlands (native connection)
3000
2200
netherlands (germany exit node)
500
350
Re: The CryptoStorm Speed Test Thread
I definitely want to help out; couple questions. Why do we avoid speedtest.net? Probably because it's a bursty test, rather than an average or median test?
Are there any particular software to track this on the user end? I know uTorrent has a sort of speed tracker over the download portion of a torrent, but outside of torrents?
Currently using the eyeball method while watching an HTTP download from a mirror.
Are there any particular software to track this on the user end? I know uTorrent has a sort of speed tracker over the download portion of a torrent, but outside of torrents?
Currently using the eyeball method while watching an HTTP download from a mirror.
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
To borrow/adapt marzametal's format:
How connection is made:
Will update with additional information as I run tests (torrents, different nodes, different mirrors)
How connection is made:
- Comcast cable; rated for 25/5, though real-world it's 30/6 on-the-wire (isn't Comcast generous?)
- Asus RT-N66U running Advanced Tomato 2.9-K131
- Wireless desktop, using 5GHz 802.11N, reads full 450Mbps connection
- Test 1: Kansas City - US Central node
- Test 2: Toyko - Japan (because Japan
)
- Windows Widget v2.22: I believe random UDP port; currently 41217. All 3 in Security tab checked
- Single download at a time (one HTTP download, one torrent, etc)
- Test 1: From University of Oklahoma mirror
- Also streaming from Digitally Imported at 128k AAC during tests (so ~130-150 kb/s with overhead); loves me some EDM
- Download via Firefox 41.0.1 between 31.2KB/s (absolute minimum observed) and 499KB/s (absolute maximum observed); also notice a dip in speed when loading up this thread, or other small (non-heavy interactive/download) sites (Slashdot (though not too light anymore), Twitter, Gmail (Ctrl+F5 refresh)
- Speeds seem to hover in the 100-250KB/sec range, jumping up and down every so often.
- Test 2: Same mirror (Univ of OK)
- Speeds as low as 5.5KB/sec, almost never getting higher than 60KB/sec
- Talk about stress-test, huh?
- Same download setup as before, just different node.
- Made DI.FM stream drop out repeatedly, while also affecting the download speed of the ISO
Will update with additional information as I run tests (torrents, different nodes, different mirrors)
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
-
- Posts: 5
- Joined: Mon Oct 12, 2015 1:53 am
Re: The CryptoStorm Speed Test Thread
I spent lots of time running tests and have generated a ton of data. Here it is:
All tests were done within about 30 minutes of each other using wget with five different servers and three different hosting providers. All were done from GNU/Linux. I am physically located on the eastern side of the United States.
Softlayer - Dallas US:
Softlayer - Seattle US:
All tests were done within about 30 minutes of each other using wget with five different servers and three different hosting providers. All were done from GNU/Linux. I am physically located on the eastern side of the United States.
Softlayer - Dallas US:
- Not Connected to CS: 29.2MB/s
- US Central: 3.03MB/s
- US West: 1.2MB/s
- Montreal: 6.90MB/s
- London: 7.23MB/s
- Switzerland: 1.04MB/s
Softlayer - Seattle US:
- Not Connected to CS: 20.3MB/s
- US Central: 2.17MB/s
- US West: 2.11MB/s
- Montreal: 7.64MB/s
- London: 5.85MB/s
- Switzerland: 1019KB/s
- Not Connected to CS: 19.0MB/s
- US Central: 2.52MB/s
- US West: 632KB/s
- Montreal: 4.6MB/s
- London: 10.8MB/s
- Switzerland: 1.2MB/s
- Not Connected to CS: 24.3MB/s
- US Central: 1.78MB/s
- US West: 1.74MB/s
- Montreal: 5.47MB/s
- London: 7.01MB/s
- Switzerland: 434KB/s
- Not Connected to CS: 28.3MB/s
- US Central: 1.73MB/s
- US West: 1.67MB/s
- Montreal: 5.84MB/s
- London: 5.97MB/s
- Switzerland: 230KB/s
Re: The CryptoStorm Speed Test Thread
@rwolslegel,@JTD121,@DonderPjem
Many thanks for posting your experiences in this thread. It would appear that routing across the United States is awful, at least for the routes used by the datacentres where Cryptostorm nodes are located. The paths to Switzerland aren't too hot either, by all accounts.
Re: speedtest.net, yes it's bursty, also browsers running a speed test written in Flash(!) aren't the most reliable things in the world. File transfer software such as FileZilla or wget (for the 'nix heads out there) gives far more accurate results.
Many thanks for posting your experiences in this thread. It would appear that routing across the United States is awful, at least for the routes used by the datacentres where Cryptostorm nodes are located. The paths to Switzerland aren't too hot either, by all accounts.
Re: speedtest.net, yes it's bursty, also browsers running a speed test written in Flash(!) aren't the most reliable things in the world. File transfer software such as FileZilla or wget (for the 'nix heads out there) gives far more accurate results.
Re: The CryptoStorm Speed Test Thread
Can't seem to edit my post above, but this is Test 3, connected to London - England node.
- Test 3: Univ of Kent UK (geographically close to London node (I assume))
- Speeds don't seem to drop as low as previous US-central node
- Speeds from 75KB/sec to 1.5MB/sec; mostly seem to fluctuate in the 150 ~ 450KB/sec range
- Prolonged idle (reading a single page of something) seems to give download a bit of a boost; observed increases between 600KB/s to 750KB/sec, several bursts 30~40 second to 1.3 - 1.5MB/sec
- Same download setup as before, just different node.
- DI.FM stream didn't skip a beat (har har!) during download
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
So with the new nodes and replacement nodes up and running, and using gigabit pipes, do we want to retest some of them?
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
@JTD121
Might be an idea. My connection is too slow to make any difference, but for those with 30Mb+ connections, it might prove useful.
EDIT:
Taken from this post:
Might be an idea. My connection is too slow to make any difference, but for those with 30Mb+ connections, it might prove useful.

EDIT:
Taken from this post:
Code: Select all
france: Speed: 1000Mb/s
frankfurt: Speed: 100Mb/s
london: Speed: 1000Mb/s
france#2: Speed: 1000Mb/s
lisbon: Speed: 1000Mb/s
seattle: Speed: 1000Mb/s
singapore: Speed: 100Mb/s
switzerland: Speed: 1000Mb/s
italy: Speed: 100Mb/s
vegas: Speed: 1000Mb/s
canada: Speed: 100Mb/s
moldova: Speed: 100Mb/s
Re: The CryptoStorm Speed Test Thread
Hm...My connection is rated about 25/5, and I've regularly seen speeds around 31/7, so I'll give this a couple test downloads soon.
I'll try to keep similar format as before (again borrowing marzametal's format).
I'll try to keep similar format as before (again borrowing marzametal's format).
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
This test site was suggested in IRC, and it looks like it's not Flash-based. Also, has an adjustable long-term speed testing method. Currently testing it out without CS to see how it fares.
Will obviously do some stress testing and compare to CS as well.
Will obviously do some stress testing and compare to CS as well.
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
Still can't seem to edit my posts above, but this is Test 4, connected to Buffalo, NY node node.
- Test 4: GOG Downloader; downloading ~830MB in total
- Speeds don't seem to fluctuate as in previous tests
- Speeds from 650KB/sec to 1.4-1.5MB/sec; Very consistent staying towards the top of this range
- Prolonged idle (reading a single page of something) doesn't seem to affect overall bandwidth.
- Also running a multi-threaded test at http://testmy.net for extra bandwidth loading
- Not really using all bandwidth available to me without CS VPN running, but definitely getting better!
- No DI.FM streaming this time
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
What about testmy.net and speedof.me for speed test usage?
Re: The CryptoStorm Speed Test Thread
@thread
Internet Connection: Cable, 50Mb/s
Local Connection: 802.11n, 72Mb/s as tested
Internet Connection: Cable, 50Mb/s
Local Connection: 802.11n, 72Mb/s as tested
Code: Select all
Node wget Torrent (avg) Torrent (peak)
Naked 5.10MB/s 4.35MiB/s 4.99MiB/s
Cantus 3.74MB/s 3.76MiB/s 5.10MiB/s
Jord 2.60MB/s 3.57MiB/s 4.43MiB/s
Tagus 2.63MB/s 3.15MiB/s 3.48MiB/s
Turing 4.58MB/s 3.88MiB/s 4.53MiB/s
Re: The CryptoStorm Speed Test Thread
Bare connection = 40Mbp/s by provider's router in modem only mode.
Local = ethernet cable - netgear switch - 500Mbp/s powerline - ASUS RT-AC68R/U (Shibby Tomato 132 running OpenVPN. Client 2 being used for this test).
Tested from UK.
----------
London.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:44:00
100%[============================================================================================================================================================>] 601,882,624 2.67MB/s in 3m 20s
2016-01-16 11:47:20 (2.87 MB/s) - ‘/dev/null’ saved [601882624/601882624]
-----------
Paris.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:48:36--
100%[============================================================================================================================================================>] 601,882,624 2.79MB/s in 4m 55s
2016-01-16 11:53:32 (1.94 MB/s) - ‘/dev/null’ saved [601882624/601882624]
----------
Frankfurt.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:56:12--
100%[============================================================================================================================================================>] 601,882,624 2.22MB/s in 4m 38s
2016-01-16 12:00:50 (2.06 MB/s) - ‘/dev/null’ saved [601882624/601882624]
-----------
Switzerland.
*$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 12:02:47--
100%[============================================================================================================================================================>] 601,882,624 2.95MB/s in 5m 39s
2016-01-16 12:08:26 (1.69 MB/s) - ‘/dev/null’ saved [601882624/601882624]
Local = ethernet cable - netgear switch - 500Mbp/s powerline - ASUS RT-AC68R/U (Shibby Tomato 132 running OpenVPN. Client 2 being used for this test).
Tested from UK.
----------
London.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:44:00
100%[============================================================================================================================================================>] 601,882,624 2.67MB/s in 3m 20s
2016-01-16 11:47:20 (2.87 MB/s) - ‘/dev/null’ saved [601882624/601882624]
-----------
Paris.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:48:36--
100%[============================================================================================================================================================>] 601,882,624 2.79MB/s in 4m 55s
2016-01-16 11:53:32 (1.94 MB/s) - ‘/dev/null’ saved [601882624/601882624]
----------
Frankfurt.
$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 11:56:12--
100%[============================================================================================================================================================>] 601,882,624 2.22MB/s in 4m 38s
2016-01-16 12:00:50 (2.06 MB/s) - ‘/dev/null’ saved [601882624/601882624]
-----------
Switzerland.
*$ wget -O /dev/null http://releases.ubuntu.com/14.04.3/ubun ... -amd64.iso
--2016-01-16 12:02:47--
100%[============================================================================================================================================================>] 601,882,624 2.95MB/s in 5m 39s
2016-01-16 12:08:26 (1.69 MB/s) - ‘/dev/null’ saved [601882624/601882624]
Re: The CryptoStorm Speed Test Thread
Strange results...
Testing by downloading http://speedtest.ams01.softlayer.com/do ... est100.zip via wget
Without VPN / Belgium:
240Mbps down / 20 Mbps up
2016-04-02 00:56:00 (25.6 MB/s) - 'test100.zip' saved [104874307/104874307]
Paris UDP:
2016-04-02 01:03:00 (259 KB/s) - 'test100.zip.1' saved [104874307/104874307]
Paris TCP:
2016-04-02 02:00:22 (2.06 MB/s) - 'test100.zip.11' saved [104874307/104874307]
Frankfurt UDP:
2016-04-02 01:18:18 (269 KB/s) - 'test100.zip.3' saved [104874307/104874307]
Frankfurt TCP:
2016-04-02 01:58:56 (1.37 MB/s) - 'test100.zip.10' saved [104874307/104874307]
Switzerland UDP:
2016-04-02 01:32:14 (267 KB/s) - 'test100.zip.5' saved [104874307/104874307]
Switzerland TCP:
2016-04-02 02:02:35 (7.21 MB/s) - 'test100.zip.12' saved [104874307/104874307]
Lisbon UDP:
2016-04-02 01:11:07 (233 KB/s) - 'test100.zip.2' saved [104874307/104874307]
Rome UDP:
2016-04-02 01:25:20 (277 KB/s) - 'test100.zip.4' saved [104874307/104874307]
Voodoo-USWest-Iceland:
2016-04-02 01:46:17 (287 KB/s) - Read error at byte 103115450/104874307 (Invalid argument). Retrying.
Voodoo-USSouth-IsleOfMan:
No results, kept reconnecting and was very unstable
Testing by downloading http://speedtest.ams01.softlayer.com/do ... est100.zip via wget
Without VPN / Belgium:
240Mbps down / 20 Mbps up
2016-04-02 00:56:00 (25.6 MB/s) - 'test100.zip' saved [104874307/104874307]
Paris UDP:
2016-04-02 01:03:00 (259 KB/s) - 'test100.zip.1' saved [104874307/104874307]
Paris TCP:
2016-04-02 02:00:22 (2.06 MB/s) - 'test100.zip.11' saved [104874307/104874307]
Frankfurt UDP:
2016-04-02 01:18:18 (269 KB/s) - 'test100.zip.3' saved [104874307/104874307]
Frankfurt TCP:
2016-04-02 01:58:56 (1.37 MB/s) - 'test100.zip.10' saved [104874307/104874307]
Switzerland UDP:
2016-04-02 01:32:14 (267 KB/s) - 'test100.zip.5' saved [104874307/104874307]
Switzerland TCP:
2016-04-02 02:02:35 (7.21 MB/s) - 'test100.zip.12' saved [104874307/104874307]
Lisbon UDP:
2016-04-02 01:11:07 (233 KB/s) - 'test100.zip.2' saved [104874307/104874307]
Rome UDP:
2016-04-02 01:25:20 (277 KB/s) - 'test100.zip.4' saved [104874307/104874307]
Voodoo-USWest-Iceland:
2016-04-02 01:46:17 (287 KB/s) - Read error at byte 103115450/104874307 (Invalid argument). Retrying.
Voodoo-USSouth-IsleOfMan:
No results, kept reconnecting and was very unstable
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
Managed a new record today, I wasn't even paying attention when I heard the fans of my device go on and checked to see what was going on. To my surprise, I was uploading at quite a nice speed, around 75 Mbps.
Connected to the Oslo node by the way.

Connected to the Oslo node by the way.

-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
New day, new record (Switzerland server): 13.26 MB/s equates to over 100 Mbit/s (106 to be precise). This is on a 200 Mbit/s line if I'm not mistaken. My ISP gave me 100 Mbit/s before but I think it has been upgraded in the meantime.


Re: The CryptoStorm Speed Test Thread
@thisis4thepeople
Many thanks for keeping us updated with your results, it's greatly appreciated. Unfortunately, I'm still stuck on an 8/1 ADSL line, so I'm unable to contribute further.
Many thanks for keeping us updated with your results, it's greatly appreciated. Unfortunately, I'm still stuck on an 8/1 ADSL line, so I'm unable to contribute further.
Re: The CryptoStorm Speed Test Thread
Well, finally I get acceptable speeds 
Constant +18MBps in and +2MBps out, this is on 350/30 line.
Switzerland node.

Constant +18MBps in and +2MBps out, this is on 350/30 line.
Switzerland node.
Re: The CryptoStorm Speed Test Thread
@turbz
Many thanks for your post, it's nice to see people still contributing to this thread. Proves that secure VPNs don't have to be slow.
Many thanks for your post, it's nice to see people still contributing to this thread. Proves that secure VPNs don't have to be slow.

-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
I'm back with new results. Last time I had a down-stream of over 13 MB/s, now it's the other way around:

Not to shabby. This was via Denmark by the way.

Not to shabby. This was via Denmark by the way.
Re: The CryptoStorm Speed Test Thread
@thisis4thepeople
Many thanks for keeping us up-to-date with throughput results, it's greatly appreciated.
If only others would chime in...
Many thanks for keeping us up-to-date with throughput results, it's greatly appreciated.

Re: The CryptoStorm Speed Test Thread
Ok, I'll bite.
40Mb bareback connection, over 5ghz wifi via Asus RT-AC68U (overclocked to 1200/800). England node. I've used the speedtest file from a post above...for some vague semblance of comparison (or I'm just lazy).
40Mb bareback connection, over 5ghz wifi via Asus RT-AC68U (overclocked to 1200/800). England node. I've used the speedtest file from a post above...for some vague semblance of comparison (or I'm just lazy).
Code: Select all
$ wget -O /dev/null http://speedtest.ams01.softlayer.com/downloads/test100.zip
100%[======================================================================================================================================>] 104,874,307 2.92MB/s in 36s
2016-09-07 09:50:29 (2.77 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Re: The CryptoStorm Speed Test Thread
@sm1th
Many thanks for sharing this. Your numbers look in line with what you connection can support.
Many thanks for sharing this. Your numbers look in line with what you connection can support.

-
- Posts: 14
- Joined: Sun Mar 22, 2015 3:25 am
Re: The CryptoStorm Speed Test Thread
Might as well..
Paris node, on a 100/10 line:
More than reasonable 
Paris node, on a 100/10 line:
Code: Select all
wget -O /dev/null http://speedtest.ams01.softlayer.com/downloads/test100.zip
HTTP request sent, awaiting response... 200 OK
Length: 104874307 (100M) [application/zip]
Saving to: ‘/dev/null’
/dev/null 100%[==============================================================================================>] 100.02M 6.42MB/s in 17s

Re: The CryptoStorm Speed Test Thread
My speeds are great. I just tested all the U.S. Servers and England and I'm getting 45down, 35up on my 50/50 connection.
I'm even getting 30down on the Isle of Man voodoo.
I'm even getting 30down on the Isle of Man voodoo.
-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
The Switzerland server is still speedy:


Re: The CryptoStorm Speed Test Thread
@thisis4thepeople
I was tempted to rent a 1Gb/s dedi and install a desktop on it just to get some numbers for my own tests, but I see there's no need. CS nodes are certainly fast enough to support the user base.
I was tempted to rent a 1Gb/s dedi and install a desktop on it just to get some numbers for my own tests, but I see there's no need. CS nodes are certainly fast enough to support the user base.

-
- Posts: 18
- Joined: Sat Jun 13, 2015 9:08 am
Re: The CryptoStorm Speed Test Thread
Switzerland server, this time from the command line with speedtest-cli.
Code: Select all
Retrieving speedtest.net configuration...
Testing from DataSource (XXX.XX.XXX.XX)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Datasource. AG (Zug) [6.26 km]: 51.064 ms
Testing download speed................................................................................
Download: 78.53 Mbit/s
Testing upload speed....................................................................................................
Upload: 39.13 Mbit/s
-
- Posts: 1
- Joined: Tue Nov 29, 2016 12:13 pm
Re: The CryptoStorm Speed Test Thread
I ran several tests from Charlotte, NC and got pretty good results on a
300Mbps cable connection routed over 802.11ac to the client
I normally get 150-200Mbps wifi speeds, so pulling 100Mbps on speedtest over VPN is pretty damn fast
large file wget download results:
(connected to the std. US East connection)
100mb file:
atlanta 3.71MB/s time 30s
newark 7.41MB/s time 13s
dallas 4.74MB/s time 21s
1000mb from Leaseweb in Virginia gave me 5.26MB/s time: 3m 1s
speedtest.net (for kicks):
100Mbps down 16.42Mbps up (up is capped at 20)
300Mbps cable connection routed over 802.11ac to the client
I normally get 150-200Mbps wifi speeds, so pulling 100Mbps on speedtest over VPN is pretty damn fast
large file wget download results:
(connected to the std. US East connection)
100mb file:
atlanta 3.71MB/s time 30s
newark 7.41MB/s time 13s
dallas 4.74MB/s time 21s
1000mb from Leaseweb in Virginia gave me 5.26MB/s time: 3m 1s
speedtest.net (for kicks):
100Mbps down 16.42Mbps up (up is capped at 20)
Re: The CryptoStorm Speed Test Thread
Here are my results:
No VPN:
cstorm_USsouth:
cstorm_USnorth:
cstorm_USeast:
cstorm_UScentral:
Code: Select all
wget -O /dev/null http://speedtest.ams01.softlayer.com/downloads/test100.zip
Code: Select all
2017-02-02 10:45:25 (3.65 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Code: Select all
2017-02-02 10:44:39 (1.07 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Code: Select all
2017-02-02 10:50:38 (1.67 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Code: Select all
2017-02-02 10:52:59 (1.85 MB/s) - ‘/dev/null’ saved [104874307/104874307]
Code: Select all
2017-02-02 10:55:52 (812 KB/s) - ‘/dev/null’ saved [104874307/104874307]
Re: The CryptoStorm Speed Test Thread
I guess it's my turn now.
How connection is being made: Duplex FTTB 25Mb/25Mb cable with cheap TP-LINK router (very stable commection. Ping between three local cities is only 9ms).
Exit nodes: Using Swiss and German exit nodes for testing.
OS: Windows 7.
Speedtest (ping/dl speed/upload speed): Original - 3ms/11.85Mbs/11.33Mbs (it's evening time, switch is probably overloaded)
Swiss node (My provider isn't hyper-restrictive, so I mainly use UDP) - 71ms/4.50Mbs/7.39Mbs (pretty decent result considering the fact that Swiss isn't that close to me)
German node (Same story - UDP, Frankfurt) - 92ms/2.92Mbs/8.87Mbs (it's LeaseWeb, so no wonder)
German node (UDP, Dusseldorf) - 97/3.8Mbs/5.04Mbs (Nothing unusual, typical German node).
I only use these as my daily driver because they're located FAIRLY close to me.
Yesterday I've had 52ms for ping and 10Mb/s for both dl/upload at Swiss node.
I usually get 1.4MB/s during torrent download. That's more than enough for me. If I'll need high speed for multimedia, I have a dedicated line and router for that.
How connection is being made: Duplex FTTB 25Mb/25Mb cable with cheap TP-LINK router (very stable commection. Ping between three local cities is only 9ms).
Exit nodes: Using Swiss and German exit nodes for testing.
OS: Windows 7.
Speedtest (ping/dl speed/upload speed): Original - 3ms/11.85Mbs/11.33Mbs (it's evening time, switch is probably overloaded)
Swiss node (My provider isn't hyper-restrictive, so I mainly use UDP) - 71ms/4.50Mbs/7.39Mbs (pretty decent result considering the fact that Swiss isn't that close to me)
German node (Same story - UDP, Frankfurt) - 92ms/2.92Mbs/8.87Mbs (it's LeaseWeb, so no wonder)
German node (UDP, Dusseldorf) - 97/3.8Mbs/5.04Mbs (Nothing unusual, typical German node).
I only use these as my daily driver because they're located FAIRLY close to me.
Yesterday I've had 52ms for ping and 10Mb/s for both dl/upload at Swiss node.
I usually get 1.4MB/s during torrent download. That's more than enough for me. If I'll need high speed for multimedia, I have a dedicated line and router for that.
Re: The CryptoStorm Speed Test Thread

Just messin' around, and used DSLReports speed test! 60s test time, 32 download/upload streams (not sure how this affects the test).
I believe this was connected to the US North or US East node cluster....
My speed tier is 200mbps down, 10mbps up. Using an Arris SB6190 with the Puma6 issues.
---------------------------------------------------------------------------------------------------
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
You derive personal satisfaction from the continued existence of the near perfect day-night cycles of the hyper cube.....
► Show Spoiler
Re: The CryptoStorm Speed Test Thread
@JTD121
Nice to see that the upstream throughput is inline with your "naked" package performance. Assuming the VPN connection starts on your PC, the downstream throughput could possibly point to congestion on that node.
Is that node the closest one to your location?
Nice to see that the upstream throughput is inline with your "naked" package performance. Assuming the VPN connection starts on your PC, the downstream throughput could possibly point to congestion on that node.
Is that node the closest one to your location?
-
- Posts: 3
- Joined: Sun May 07, 2017 5:29 am
Re: The CryptoStorm Speed Test Thread
Connection: 4G
Exit Node: England
Exit Node: England
Code: Select all
wget http://cachefly.cachefly.net/100mb.test
--2017-06-02 04:19:36-- http://cachefly.cachefly.net/100mb.test
Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175
Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: ‘100mb.test.2’
100%[================================================================================================>] 104,857,600 3.36MB/s in 31s
2017-06-02 04:20:12 (3.25 MB/s) - ‘100mb.test.2’ saved [104857600/104857600]
Re: The CryptoStorm Speed Test Thread
Post upgrade 'ECC':
VPN connected to ASUS RT-AC68U ('FreshTomato' firmware) overclocked to 1200/800.
OpenVPN 2.4.6, OpenSSL 1.0.2p
Base connection is ~40 Mbits down, 10 Mbits up. VDSL.
Connected to closest exit node.
Test was conducted on an old (and much loved) thinkpad connected to router via ethernet. I'm not bothering with a wifi test, too many variables for it to mean much imho.
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 14.871 ms
Testing download speed................................................................................
Download: 33.77 Mbit/s
Testing upload speed................................................................................................
Upload: 8.95 Mbit/s
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 15.385 ms
Testing download speed................................................................................
Download: 34.43 Mbit/s
Testing upload speed................................................................................................
Upload: 8.93 Mbit/s
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 15.958 ms
Testing download speed................................................................................
Download: 34.74 Mbit/s
Testing upload speed................................................................................................
Upload: 8.86 Mbit/s
I thought the new fancy ways would be a bit slower, but it seems not.
VPN connected to ASUS RT-AC68U ('FreshTomato' firmware) overclocked to 1200/800.
OpenVPN 2.4.6, OpenSSL 1.0.2p
Base connection is ~40 Mbits down, 10 Mbits up. VDSL.
Connected to closest exit node.
Test was conducted on an old (and much loved) thinkpad connected to router via ethernet. I'm not bothering with a wifi test, too many variables for it to mean much imho.
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 14.871 ms
Testing download speed................................................................................
Download: 33.77 Mbit/s
Testing upload speed................................................................................................
Upload: 8.95 Mbit/s
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 15.385 ms
Testing download speed................................................................................
Download: 34.43 Mbit/s
Testing upload speed................................................................................................
Upload: 8.93 Mbit/s
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from UK Dedicated Servers Limited (5.101.149.4)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Uno (Milton Keynes) [49.22 km]: 15.958 ms
Testing download speed................................................................................
Download: 34.74 Mbit/s
Testing upload speed................................................................................................
Upload: 8.86 Mbit/s
I thought the new fancy ways would be a bit slower, but it seems not.

Re: The CryptoStorm Speed Test Thread
@blurb
Many thanks for sharing this.
Ironically, while the newer ECC cryptographic algorithms are much more efficient than the older RSA - requiring much less CPU power for the same level of cryptographic strength - there's a good chance that the older routers (which would actually benefit from this) will never be updated to take advantage of it.
Many thanks for sharing this.

Ironically, while the newer ECC cryptographic algorithms are much more efficient than the older RSA - requiring much less CPU power for the same level of cryptographic strength - there's a good chance that the older routers (which would actually benefit from this) will never be updated to take advantage of it.
Re: The CryptoStorm Speed Test Thread
Question: Is a cheap vps set up using https://github.com/StreisandEffect/streisand faster than Cryptostorm? To make it a bit unfair, I'm connecting directly from my laptop and not the router.
Comparison is to my last post in this thread.
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from Linode (***.***.***.***)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by 23media GmbH (Frankfurt) [0.53 km]: 31.704 ms
Testing download speed.........................................................
Download: 34.58 Mbit/s
Testing upload speed...........................................................
Upload: 8.87 Mbit/s
Answer: Not really, no, not on my old laptop.
Just thought I'd share.
Comparison is to my last post in this thread.
$ ./speedtest-cli
Retrieving speedtest.net configuration...
Testing from Linode (***.***.***.***)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by 23media GmbH (Frankfurt) [0.53 km]: 31.704 ms
Testing download speed.........................................................
Download: 34.58 Mbit/s
Testing upload speed...........................................................
Upload: 8.87 Mbit/s
Answer: Not really, no, not on my old laptop.
Just thought I'd share.