Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links


Categories

In this Discussion

› › GreenBench Benchmark Script - Share the results - Report bugs - Request new features - Page 2
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

GreenBench Benchmark Script - Share the results - Report bugs - Request new features

2»

Comments

  • NDTNNDTN Member, Patron Provider, Top Host

    @AXYZE said:
    That's gonna be harsh take from me.

    This script is inspired from other scripts (YABS, Tocdo.io, bench.sh) but with more features such as URL Sharing, options to choose iperf or speedtest-cli

    This tool has 1:1 output style as 'YABS', its clearly starting point for your project.

    @Liso did the same with his Benchy

    That's okay. Whats not okay is forking open source project (yabs), making new improvements (which were already done by @Liso - both Speedtest & URL sharing), then slapping your commercial logo and close sourcing it by providing just a binary file.

    I honestly hate when companies just use open source work without contributing anything. In this case its not just "using open source". It even has 1:1 output style as 'yabs'.
    Its not 'inspired'. Fake iPhone is also "inspired by Apple"?

    If Mason & Liso are okay with this then ok, I cant really compain.
    But if you didn't even asked both of them then... some bad words.

    No matter if you asked them I don't support it. Why?

    1. We have open source tools that work good. Going closed source is going backward.
    2. You used different providers servers for speedtest instead of your own servers (seriously? did you ask them about it or you just use their bandwidth without asking for your own tool where you promote own 'green' brand?). Real improvement would be communicating with own servers to check if they have available slots to test just to be sure that these bandwidth numbers are legit. You could also shuffle between them when they would have available slots instead of going 1 2 3.
    3. You are provider. Your business is selling VPS and now you promote own tool to evaluate these VPSes. Imagine Xiaomi making benchmark to test Android phones - they could just fake these results for own products or make it worse for main competitiors. Its closed source, nobody will get an idea. Even if it would be opensourced you could just use benchmark settings that would favor own devices. I'm not telling that you could done these exact things, but this tool is conflicting with your business. Reviews and tests should be done be reviewers and testers, not companies that make the product. Same with tools that are used for these reviews and tests.

    I'm completely against it. Sorry.

    Thank you for your feedback. I would like to clear several things here:

    1. We do not deny the fact that the GreenBench script was written based on YABS and several other benchmark scripts, it's stated on our site (https://bench.green.cloud/), let me know if you think it's not clear enough.
    2. I'm not sure if @Liso has Speedtest & URL sharing features, let me double check it again. But the Speedtest & URL sharing ideas were taken from other scripts (tocdo.io & bench.sh & bench.monster). Our original idea to build this script is to fill the gap between benchmark scripts, combining the features that users need.
    3. The reason why we encrypted the script was explained on my previous comment.
    4. I dont think it's an issue using public test servers. It would be an issue if we charge for using our script and using public servers, but it's not the case here. We will gradually change the test servers, it was my requirement for our dev at the beginning but it would take more time so I asked him to make the script works first then im
    5. We can ask someone that can audit the script if you think we will fake the results, a LET admin maybe. But if you dont trust or support it, then dont use it. Comparing between providers is just an idea now, if most of the users find it unfair/unnecessary then we will not implement that.
    6. The script is still under development, that's why we have this thread. We build this script from our own needs, as in the past we have been asked multiple times for YABS or benchmark and it took a lot of time for searching and sharing the results. It's now easier for us to do that with GreenBench script. So I think why not making it public so it might be useful for other users.
  • Peppery9Peppery9 Member

    Wow, it is actually submitting results via FTP.

    Thanked by 3AXYZE devp Daniel15
  • NDTNNDTN Member, Patron Provider, Top Host

    @Peppery9 said:

    Wow, it is actually submitting results via FTP.

    Yes to a public folder. We are still trying to find a better way to do that.

  • Peppery9Peppery9 Member

    @NDTN said:
    Yes to a public folder. We are still trying to find a better way to do that.

    Use cURL to upload your results. See how @Liso is doing it: https://github.com/L1so/benchy/blob/main/benchy#L122

    Thanked by 1NDTN
  • NDTNNDTN Member, Patron Provider, Top Host

    @Peppery9 said:

    @NDTN said:
    Yes to a public folder. We are still trying to find a better way to do that.

    Use cURL to upload your results. See how @Liso is doing it: https://github.com/L1so/benchy/blob/main/benchy#L122

    Thanks! I will ask our dev to take a look at that.

  • AXYZEAXYZE Member
    edited June 2022

    @NDTN said:

    @AXYZE said:
    That's gonna be harsh take from me.

    This script is inspired from other scripts (YABS, Tocdo.io, bench.sh) but with more features such as URL Sharing, options to choose iperf or speedtest-cli

    This tool has 1:1 output style as 'YABS', its clearly starting point for your project.

    @Liso did the same with his Benchy

    That's okay. Whats not okay is forking open source project (yabs), making new improvements (which were already done by @Liso - both Speedtest & URL sharing), then slapping your commercial logo and close sourcing it by providing just a binary file.

    I honestly hate when companies just use open source work without contributing anything. In this case its not just "using open source". It even has 1:1 output style as 'yabs'.
    Its not 'inspired'. Fake iPhone is also "inspired by Apple"?

    If Mason & Liso are okay with this then ok, I cant really compain.
    But if you didn't even asked both of them then... some bad words.

    No matter if you asked them I don't support it. Why?

    1. We have open source tools that work good. Going closed source is going backward.
    2. You used different providers servers for speedtest instead of your own servers (seriously? did you ask them about it or you just use their bandwidth without asking for your own tool where you promote own 'green' brand?). Real improvement would be communicating with own servers to check if they have available slots to test just to be sure that these bandwidth numbers are legit. You could also shuffle between them when they would have available slots instead of going 1 2 3.
    3. You are provider. Your business is selling VPS and now you promote own tool to evaluate these VPSes. Imagine Xiaomi making benchmark to test Android phones - they could just fake these results for own products or make it worse for main competitiors. Its closed source, nobody will get an idea. Even if it would be opensourced you could just use benchmark settings that would favor own devices. I'm not telling that you could done these exact things, but this tool is conflicting with your business. Reviews and tests should be done be reviewers and testers, not companies that make the product. Same with tools that are used for these reviews and tests.

    I'm completely against it. Sorry.

    Thank you for your feedback. I would like to clear several things here:

    1. We do not deny the fact that the GreenBench script was written based on YABS and several other benchmark scripts, it's stated on our site (https://bench.green.cloud/), let me know if you think it's not clear enough.
    2. I'm not sure if @Liso has Speedtest & URL sharing features, let me double check it again. But the Speedtest & URL sharing ideas were taken from other scripts (tocdo.io & bench.sh & bench.monster). Our original idea to build this script is to fill the gap between benchmark scripts, combining the features that users need.
    3. The reason why we encrypted the script was explained on my previous comment.
    4. I dont think it's an issue using public test servers. It would be an issue if we charge for using our script and using public servers, but it's not the case here. We will gradually change the test servers, it was my requirement for our dev at the beginning but it would take more time so I asked him to make the script works first then im
    5. We can ask someone that can audit the script if you think we will fake the results, a LET admin maybe. But if you dont trust or support it, then dont use it. Comparing between providers is just an idea now, if most of the users find it unfair/unnecessary then we will not implement that.
    6. The script is still under development, that's why we have this thread. We build this script from our own needs, as in the past we have been asked multiple times for YABS or benchmark and it took a lot of time for searching and sharing the results. It's now easier for us to do that with GreenBench script. So I think why not making it public so it might be useful for other users.
    1. They are open source. Did you ask them if you can use their work to make your own closed source product?
    2. You dont need to double check, here's Benchy URL sharing feature http://sprunge.us/3FZtpY
    3. That just sounds like an excuse, sorry. Benchmark script doesn't need to be encrypted to upload text output to server. Even if you do it via FTP (lol?). Permissions exists. Validation exists. You can't write script without knowing these things, especially if you manipulate output of iperf in order to put them into nice table. Make the same thing but for website - send just a couple text values to server and let the server make nice output out of it.
    4. So you're telling me I can blast several Gbps from several sources into your open 'public servers' (lets say looking glass one, its public and often has 'speedtest' files) without asking and I can brand this as my own 'benchmark'? ... okay. I'm just curious how other providers look at this, so I'll wait for other replies.
    5. I never said you fake results, I said that there's such possibility (faking OR changing benchmark values in order to favor your services) because your business is selling VPSes and you built tool to test them and compare them across different providers.
    6. You just paste the same 'curl' command and you copy URL instead of output. You didn't need to build whole new tool when we have well tested and open source 'yabs' and people ask you to test 'yabs'. If you need to manage results of 'yabs' there's already 'My idlers', but okay... you could built it in-house in a more convenient way for you. What I'm saying is that puting your own brand on benchmark product instead of using 'yabs' (which is well known for long time) decreases credibility. You cant deny that. Once again, people ask for 'yabs', not own tool which is benchmarking own product and comparing it to competitors. Other companies are using 'yabs' even in their docs for example https://webdock.io/en/docs/webdock-control-panel/optimizing-performance/performance#ssd-pro
    Thanked by 2devp jugganuts
  • guagua_ya66guagua_ya66 Member

    Virmach Ryzen JP

    root@debian:~# bash <(wget -qO- https://green.cloud/install)
* ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** *
*              GreenBench Scripts v1.0.2             *
*                  Update: 31-05-2022                *
*              https://bench.green.cloud             *
* ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** *

Thu 02 Jun 2022 06:19:57 PM CST

Basic System Information by GreenBench:
---------------------------------
Processor      : AMD Ryzen 9 5950X 16-Core Processor
CPU cores      : 3 core(s) @ 3393.624 MHz
AES-NI         : ✔ Enabled
VM-x/AMD-V     : ✔ Enabled
Virtualization : KVM
OS             : Debian GNU/Linux 11 (bullseye)
Total Disk     : 77.2 GB (943.2 MB Used)
Total Mem      : 3.8 GB (64.3 MB Used)
Total Swap     : 975.0 MB (0 Used)
Uptime         : 18:19:57 up 1 day,  2:03,  1 user,  load average: 0.00, 0.00, 0.00
Organization   : AS23959 Owl Limited
Location       : Tokyo / JP
Region         : Tokyo

Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | Read         | Write       | Total       | IOPS Read  | IOPS Write | IOPS
──────     | ─────        | ─────       | ─────       | ─────      | ─────      | ─────
4k         | 328.73 MB/s  | 329.60 MB/s | 658.34 MB/s | 82.1k      | 82.4k      | 164.5k
------     | ---          | ----        | ----        | ----       | ----       | ----
64k        | 1.21 GB/s    | 1.22 GB/s   | 2.44 GB/s   | 19.0k      | 19.1k      | 38.1k
------     | ---          | ----        | ----        | ----       | ----       | ----
512k       | 2.84 GB/s    | 2.99 GB/s   | 5.84 GB/s   | 5.5k       | 5.8k       | 11.4k
------     | ---          | ----        | ----        | ----       | ----       | ----
1m         | 3.09 GB/s    | 3.29 GB/s   | 6.39 GB/s   | 3.0k       | 3.2k       | 6.2k

Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed
  ------        | ---                       | ----            | ----
Clouvider       | London, UK (10G)          | 346 Mbits/sec   | 152 Mbits/sec
Online.net      | Paris, FR (10G)           | 513 Mbits/sec   | 135 Mbits/sec
WorldStream     | The Netherlands (10G)     | 385 Mbits/sec   | 188 Mbits/sec
WebHorizon      | Singapore (400M)          | 600 Mbits/sec   | busy
Clouvider       | NYC, NY, US (10G)         | 459 Mbits/sec   | 211 Mbits/sec
Clouvider       | Los Angeles, CA, US (10G) | 555 Mbits/sec   | 134 Mbits/sec
Clouvider       | Frankfurt, DE (1G)        | 399 Mbits/sec   | 152 Mbits/sec

Geekbench 5 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 933
Multi Core      | 2181
Full Test       | https://browser.geekbench.com/v5/cpu/15254463

Share the result: https://bench.green.cloud/4a3b3d1c08f2b977ac94fed5bcc105c027274
  • Peppery9Peppery9 Member
    edited June 2022

    @NDTN said:

    @Peppery9 said:

    @NDTN said:
    Yes to a public folder. We are still trying to find a better way to do that.

    Use cURL to upload your results. See how @Liso is doing it: https://github.com/L1so/benchy/blob/main/benchy#L122

    Thanks! I will ask our dev to take a look at that.

    I highly recommend you take the FTP server offline while this is fixed.

    Thanked by 7caracal bulbasaur NDTN dev077 ariq01 Xrmaddness devp
  • NDTNNDTN Member, Patron Provider, Top Host

    @AXYZE said:

    @NDTN said:

    @AXYZE said:
    That's gonna be harsh take from me.

    This script is inspired from other scripts (YABS, Tocdo.io, bench.sh) but with more features such as URL Sharing, options to choose iperf or speedtest-cli

    This tool has 1:1 output style as 'YABS', its clearly starting point for your project.

    @Liso did the same with his Benchy

    That's okay. Whats not okay is forking open source project (yabs), making new improvements (which were already done by @Liso - both Speedtest & URL sharing), then slapping your commercial logo and close sourcing it by providing just a binary file.

    I honestly hate when companies just use open source work without contributing anything. In this case its not just "using open source". It even has 1:1 output style as 'yabs'.
    Its not 'inspired'. Fake iPhone is also "inspired by Apple"?

    If Mason & Liso are okay with this then ok, I cant really compain.
    But if you didn't even asked both of them then... some bad words.

    No matter if you asked them I don't support it. Why?

    1. We have open source tools that work good. Going closed source is going backward.
    2. You used different providers servers for speedtest instead of your own servers (seriously? did you ask them about it or you just use their bandwidth without asking for your own tool where you promote own 'green' brand?). Real improvement would be communicating with own servers to check if they have available slots to test just to be sure that these bandwidth numbers are legit. You could also shuffle between them when they would have available slots instead of going 1 2 3.
    3. You are provider. Your business is selling VPS and now you promote own tool to evaluate these VPSes. Imagine Xiaomi making benchmark to test Android phones - they could just fake these results for own products or make it worse for main competitiors. Its closed source, nobody will get an idea. Even if it would be opensourced you could just use benchmark settings that would favor own devices. I'm not telling that you could done these exact things, but this tool is conflicting with your business. Reviews and tests should be done be reviewers and testers, not companies that make the product. Same with tools that are used for these reviews and tests.

    I'm completely against it. Sorry.

    Thank you for your feedback. I would like to clear several things here:

    1. We do not deny the fact that the GreenBench script was written based on YABS and several other benchmark scripts, it's stated on our site (https://bench.green.cloud/), let me know if you think it's not clear enough.
    2. I'm not sure if @Liso has Speedtest & URL sharing features, let me double check it again. But the Speedtest & URL sharing ideas were taken from other scripts (tocdo.io & bench.sh & bench.monster). Our original idea to build this script is to fill the gap between benchmark scripts, combining the features that users need.
    3. The reason why we encrypted the script was explained on my previous comment.
    4. I dont think it's an issue using public test servers. It would be an issue if we charge for using our script and using public servers, but it's not the case here. We will gradually change the test servers, it was my requirement for our dev at the beginning but it would take more time so I asked him to make the script works first then im
    5. We can ask someone that can audit the script if you think we will fake the results, a LET admin maybe. But if you dont trust or support it, then dont use it. Comparing between providers is just an idea now, if most of the users find it unfair/unnecessary then we will not implement that.
    6. The script is still under development, that's why we have this thread. We build this script from our own needs, as in the past we have been asked multiple times for YABS or benchmark and it took a lot of time for searching and sharing the results. It's now easier for us to do that with GreenBench script. So I think why not making it public so it might be useful for other users.
    1. They are open source. Did you ask them if you can use their work to make your own closed source product?
    2. You dont need to double check, here's Benchy URL sharing feature http://sprunge.us/3FZtpY
    3. That just sounds like an excuse, sorry. Benchmark script doesn't need to be encrypted to upload text output to server. Even if you do it via FTP (lol?). Permissions exists. Validation exists. You can't write script without knowing these things, especially if you manipulate output of iperf in order to put them into nice table. Make the same thing but for website - send just a couple text values to server and let the server make nice output out of it.
    4. So you're telling me I can blast several Gbps from several sources into your open 'public servers' (lets say looking glass one, its public and often has 'speedtest' files) without asking and I can brand this as my own 'benchmark'? ... okay. I'm just curious how other providers look at this, so I'll wait for other replies.
    5. I never said you fake results, I said that there's such possibility (faking OR changing benchmark values in order to favor your services) because your business is selling VPSes and you built tool to test them and compare them across different providers.
    6. You just paste the same 'curl' command and you copy URL instead of output. You didn't need to build whole new tool when we have well tested and open source 'yabs' and people ask you to test 'yabs'. If you need to manage results of 'yabs' there's already 'My idlers', but okay... you could built it in-house in a more convenient way for you. What I'm saying is that puting your own brand on benchmark product instead of using 'yabs' (which is well known for long time) decreases credibility. You cant deny that. Once again, people ask for 'yabs', not own tool which is benchmarking own product and comparing it to competitors.

    1, 2, 3: I did not know that Benchy has URL sharing feature. I'm not a dev so I cant comment on complex technical concerns. But I will ask our dev about that and make sure to have the open source version of our script.
    4. Not an excuse. As said, we have already had the plan to implement our own test servers, I will contact the other providers to ask for the permission to use the public test servers.
    5. This can be resolved when we make the script open source?
    6. The script is useful for us and we think it might be useful for other users, if not then we will just use it privately, no issue here.

    We will take the script offline to fix the reported bugs. I will update this thread once I have more information from the dev.

  • bulbasaurbulbasaur Member

    @NDTN, would you please answer why this is a compiled binary?

  • corbpiecorbpie Member

    @Peppery9 said:

    Wow, it is actually submitting results via FTP.

    Genius dev

    Thanked by 2AXYZE Xrmaddness
  • MallocVoidstarMallocVoidstar Member

    @AXYZE said:

    @NDTN said:

    @AXYZE said:
    That's gonna be harsh take from me.

    This script is inspired from other scripts (YABS, Tocdo.io, bench.sh) but with more features such as URL Sharing, options to choose iperf or speedtest-cli

    This tool has 1:1 output style as 'YABS', its clearly starting point for your project.

    @Liso did the same with his Benchy

    That's okay. Whats not okay is forking open source project (yabs), making new improvements (which were already done by @Liso - both Speedtest & URL sharing), then slapping your commercial logo and close sourcing it by providing just a binary file.

    I honestly hate when companies just use open source work without contributing anything. In this case its not just "using open source". It even has 1:1 output style as 'yabs'.
    Its not 'inspired'. Fake iPhone is also "inspired by Apple"?

    If Mason & Liso are okay with this then ok, I cant really compain.
    But if you didn't even asked both of them then... some bad words.

    No matter if you asked them I don't support it. Why?

    1. We have open source tools that work good. Going closed source is going backward.
    2. You used different providers servers for speedtest instead of your own servers (seriously? did you ask them about it or you just use their bandwidth without asking for your own tool where you promote own 'green' brand?). Real improvement would be communicating with own servers to check if they have available slots to test just to be sure that these bandwidth numbers are legit. You could also shuffle between them when they would have available slots instead of going 1 2 3.
    3. You are provider. Your business is selling VPS and now you promote own tool to evaluate these VPSes. Imagine Xiaomi making benchmark to test Android phones - they could just fake these results for own products or make it worse for main competitiors. Its closed source, nobody will get an idea. Even if it would be opensourced you could just use benchmark settings that would favor own devices. I'm not telling that you could done these exact things, but this tool is conflicting with your business. Reviews and tests should be done be reviewers and testers, not companies that make the product. Same with tools that are used for these reviews and tests.

    I'm completely against it. Sorry.

    Thank you for your feedback. I would like to clear several things here:

    1. We do not deny the fact that the GreenBench script was written based on YABS and several other benchmark scripts, it's stated on our site (https://bench.green.cloud/), let me know if you think it's not clear enough.
    2. I'm not sure if @Liso has Speedtest & URL sharing features, let me double check it again. But the Speedtest & URL sharing ideas were taken from other scripts (tocdo.io & bench.sh & bench.monster). Our original idea to build this script is to fill the gap between benchmark scripts, combining the features that users need.
    3. The reason why we encrypted the script was explained on my previous comment.
    4. I dont think it's an issue using public test servers. It would be an issue if we charge for using our script and using public servers, but it's not the case here. We will gradually change the test servers, it was my requirement for our dev at the beginning but it would take more time so I asked him to make the script works first then im
    5. We can ask someone that can audit the script if you think we will fake the results, a LET admin maybe. But if you dont trust or support it, then dont use it. Comparing between providers is just an idea now, if most of the users find it unfair/unnecessary then we will not implement that.
    6. The script is still under development, that's why we have this thread. We build this script from our own needs, as in the past we have been asked multiple times for YABS or benchmark and it took a lot of time for searching and sharing the results. It's now easier for us to do that with GreenBench script. So I think why not making it public so it might be useful for other users.
    1. They are open source. Did you ask them if you can use their work to make your own closed source product?

    Specifically regarding this: YABS is licensed under the WTFPL. My understanding is that taking its code and making it closed source without credit or requesting permission is in fact allowed under its terms ("You Just DO WHAT THE FUCK YOU WANT TO").

  • AXYZEAXYZE Member
    edited June 2022

    @MallocVoidstar said:
    Specifically regarding this: YABS is licensed under the WTFPL. My understanding is that taking its code and making it closed source without credit or requesting permission is in fact allowed under its terms ("You Just DO WHAT THE FUCK YOU WANT TO").

    Thats right, thus I dont say they violated some license :)
    just curious if they even asked as I personally hate when companies use open source products to make own closed source products with 0 contributions.
    I would never raise this question if it would be open source, such forks are great!

    It would be even better if they would just make pure fork of 'yabs' like 'Yabs Green' or made pull request with added features. Its the same thing, why inventing it once again with more limitations (GreenBench doesnt even work on ARM) when there's working open source product available and you can CONTRIBUTE? :)

    Thanked by 1devp
  • NoCommentNoComment Member
    edited June 2022

    @AXYZE said:

    @MallocVoidstar said:
    Specifically regarding this: YABS is licensed under the WTFPL. My understanding is that taking its code and making it closed source without credit or requesting permission is in fact allowed under its terms ("You Just DO WHAT THE FUCK YOU WANT TO").

    Thats right, thus I dont say they violated some license :)
    just curious if they even asked as I personally hate when companies use open source products to make own closed source products with 0 contributions.
    I would never raise this question if it would be open source, such forks are great!

    It would be even better if they would just make pure fork of 'yabs' like 'Yabs Green' or made pull request with added features. Its the same thing, why inventing it once again with more limitations (GreenBench doesnt even work on ARM) when there's working open source product available and you can CONTRIBUTE? :)

    For other scripts like bench monster, it is a MIT license so if their code is not open-source, technically they have to credit the used portions of code properly.

    Anyway, I would not run a compiled binary in a server even if it is probably safe just to run some benchmark script.

    Thanked by 1devp
  • AXYZEAXYZE Member
    edited June 2022

    @NDTN said:

    @AXYZE said:

    @NDTN said:

    @AXYZE said:
    That's gonna be harsh take from me.

    This script is inspired from other scripts (YABS, Tocdo.io, bench.sh) but with more features such as URL Sharing, options to choose iperf or speedtest-cli

    This tool has 1:1 output style as 'YABS', its clearly starting point for your project.

    @Liso did the same with his Benchy

    That's okay. Whats not okay is forking open source project (yabs), making new improvements (which were already done by @Liso - both Speedtest & URL sharing), then slapping your commercial logo and close sourcing it by providing just a binary file.

    I honestly hate when companies just use open source work without contributing anything. In this case its not just "using open source". It even has 1:1 output style as 'yabs'.
    Its not 'inspired'. Fake iPhone is also "inspired by Apple"?

    If Mason & Liso are okay with this then ok, I cant really compain.
    But if you didn't even asked both of them then... some bad words.

    No matter if you asked them I don't support it. Why?

    1. We have open source tools that work good. Going closed source is going backward.
    2. You used different providers servers for speedtest instead of your own servers (seriously? did you ask them about it or you just use their bandwidth without asking for your own tool where you promote own 'green' brand?). Real improvement would be communicating with own servers to check if they have available slots to test just to be sure that these bandwidth numbers are legit. You could also shuffle between them when they would have available slots instead of going 1 2 3.
    3. You are provider. Your business is selling VPS and now you promote own tool to evaluate these VPSes. Imagine Xiaomi making benchmark to test Android phones - they could just fake these results for own products or make it worse for main competitiors. Its closed source, nobody will get an idea. Even if it would be opensourced you could just use benchmark settings that would favor own devices. I'm not telling that you could done these exact things, but this tool is conflicting with your business. Reviews and tests should be done be reviewers and testers, not companies that make the product. Same with tools that are used for these reviews and tests.

    I'm completely against it. Sorry.

    Thank you for your feedback. I would like to clear several things here:

    1. We do not deny the fact that the GreenBench script was written based on YABS and several other benchmark scripts, it's stated on our site (https://bench.green.cloud/), let me know if you think it's not clear enough.
    2. I'm not sure if @Liso has Speedtest & URL sharing features, let me double check it again. But the Speedtest & URL sharing ideas were taken from other scripts (tocdo.io & bench.sh & bench.monster). Our original idea to build this script is to fill the gap between benchmark scripts, combining the features that users need.
    3. The reason why we encrypted the script was explained on my previous comment.
    4. I dont think it's an issue using public test servers. It would be an issue if we charge for using our script and using public servers, but it's not the case here. We will gradually change the test servers, it was my requirement for our dev at the beginning but it would take more time so I asked him to make the script works first then im
    5. We can ask someone that can audit the script if you think we will fake the results, a LET admin maybe. But if you dont trust or support it, then dont use it. Comparing between providers is just an idea now, if most of the users find it unfair/unnecessary then we will not implement that.
    6. The script is still under development, that's why we have this thread. We build this script from our own needs, as in the past we have been asked multiple times for YABS or benchmark and it took a lot of time for searching and sharing the results. It's now easier for us to do that with GreenBench script. So I think why not making it public so it might be useful for other users.
    1. They are open source. Did you ask them if you can use their work to make your own closed source product?
    2. You dont need to double check, here's Benchy URL sharing feature http://sprunge.us/3FZtpY
    3. That just sounds like an excuse, sorry. Benchmark script doesn't need to be encrypted to upload text output to server. Even if you do it via FTP (lol?). Permissions exists. Validation exists. You can't write script without knowing these things, especially if you manipulate output of iperf in order to put them into nice table. Make the same thing but for website - send just a couple text values to server and let the server make nice output out of it.
    4. So you're telling me I can blast several Gbps from several sources into your open 'public servers' (lets say looking glass one, its public and often has 'speedtest' files) without asking and I can brand this as my own 'benchmark'? ... okay. I'm just curious how other providers look at this, so I'll wait for other replies.
    5. I never said you fake results, I said that there's such possibility (faking OR changing benchmark values in order to favor your services) because your business is selling VPSes and you built tool to test them and compare them across different providers.
    6. You just paste the same 'curl' command and you copy URL instead of output. You didn't need to build whole new tool when we have well tested and open source 'yabs' and people ask you to test 'yabs'. If you need to manage results of 'yabs' there's already 'My idlers', but okay... you could built it in-house in a more convenient way for you. What I'm saying is that puting your own brand on benchmark product instead of using 'yabs' (which is well known for long time) decreases credibility. You cant deny that. Once again, people ask for 'yabs', not own tool which is benchmarking own product and comparing it to competitors.

    1, 2, 3: I did not know that Benchy has URL sharing feature. I'm not a dev so I cant comment on complex technical concerns. But I will ask our dev about that and make sure to have the open source version of our script.
    4. Not an excuse. As said, we have already had the plan to implement our own test servers, I will contact the other providers to ask for the permission to use the public test servers.
    5. This can be resolved when we make the script open source?
    6. The script is useful for us and we think it might be useful for other users, if not then we will just use it privately, no issue here.

    We will take the script offline to fix the reported bugs. I will update this thread once I have more information from the dev.

    1. What about completely different approach?

    Contribute to Yabs
    Contact Mason (yabs creator) about URL sharing feature. He can add it to 'yabs', you just need to make an very easy API (10 minute work) and make one small VPS to host it.
    Place small logo on results page (hosted on GreenCloud).
    You will receive a lot of views, brand recognition (yabs is used by a lot of people)

    Make open source fork
    If Mason doesn't agree then make your own 'yabs green'. ADD features instead of making it from scratch and encrypting it for some FTP upload (lol). Small, easy API is enough to upload results.

    Whatever you choose you wont be in charge of actually making the code which benchmarks your own services (conflict of interests), it stays opensource, works everywhere (ARM too) and its easier to promote. Zero cons.

    Thanked by 2bulbasaur devp
  • corbpiecorbpie Member

    This seems like a copy and paste job, was it written from nothing or did it start as a base from one of the inspiration benches?

  • dIsKdIsK Member

    submitting results via ftp? :o bro please :(

  • Peppery9Peppery9 Member
    edited June 2022

    @corbpie said:
    This seems like a copy and paste job, was it written from nothing or did it start as a base from one of the inspiration benches?

    It's based on yabs:

    Warning! You are running GreenBench on a ZFS Filesystem and your disk space is too low for the fio test. Your test results will be inaccurate. You need at least 48 GB free in order to complete this test accurately. For more information, please see https://github.com/masonr/yet-another-bench-script/issues/13

    If it were based on Benchy we could've avoided the FTP security nightmare.

    Fundamentally I see nothing wrong with a fork of another benchmark script. However piping scripts direct to bash is awful practice at the best of times, doing it with a closed binary is even worse.

    Thanked by 2bulbasaur devp
  • devpdevp Member
    edited June 2022

    @Peppery9 said:

    @corbpie said:
    This seems like a copy and paste job, was it written from nothing or did it start as a base from one of the inspiration benches?

    It's based on yabs:

    Warning! You are running GreenBench on a ZFS Filesystem and your disk space is too low for the fio test. Your test results will be inaccurate. You need at least 48 GB free in order to complete this test accurately. For more information, please see https://github.com/masonr/yet-another-bench-script/issues/13

    If it were based on Benchy we could've avoided the FTP security nightmare.

    Fundamentally I see nothing wrong with a fork of another benchmark script. However piping scripts direct to bash is awful practice at the best of times, doing it with a closed binary is even worse.

    They copy pasted open source stuff and build a wrapper script around than compile that to binary having their tag and calling that open source work of other contributors as their proprietary stuff.

    This implementation of wrapper script is posted and promoted by a hosting provider and they are giving random reasons.

    Thanked by 1AXYZE
  • AXYZEAXYZE Member

    @Peppery9 said:

    @corbpie said:
    This seems like a copy and paste job, was it written from nothing or did it start as a base from one of the inspiration benches?

    It's based on yabs:

    Warning! You are running GreenBench on a ZFS Filesystem and your disk space is too low for the fio test. Your test results will be inaccurate. You need at least 48 GB free in order to complete this test accurately. For more information, please see https://github.com/masonr/yet-another-bench-script/issues/13

    If it were based on Benchy we could've avoided the FTP security nightmare.

    Fundamentally I see nothing wrong with a fork of another benchmark script. However piping scripts direct to bash is awful practice at the best of times, doing it with a closed binary is even worse.

    I thought it cant get worse... This really needs an explaination.

    Original post states "This script is inspired from other scripts (YABS, Tocdo.io, bench.sh)".
    Is copy&pasting someone else work a inspiration? 😵

  • corbpiecorbpie Member

    The under construction page was "inspired" from w3 schools coming soon page

    Thanked by 4hyperblast devp bulbasaur Daniel15
  • hyperblasthyperblast Member

    https://bench.green.cloud/ - UNDER CONSTRUCTION

  • jugganutsjugganuts Member

    I'm with @AXYZE this benchmark script is pointless and a backstep. We should be supporting and making pull requests for @Liso and @MasonR s scripts. Not copy and pasted code and closed binaries with questionable practices...

    Thanked by 1bulbasaur
  • raviravi Member

    Now the site is using a pre-made template for under construction page:
    https://www.w3schools.com/howto/howto_css_coming_soon.asp

  • larmaratlarmarat Member

    Wasn't the worst marketing plan originally but this is a shitshow

  • lowestlowest Member
    edited September 2023 
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2023-09-06                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Fri Sep  8 10:07:44 PM BST 2023

Basic System Information:
---------------------------------
Uptime     : 0 days, 1 hours, 8 minutes
Processor  : AMD EPYC 7763 64-Core Processor
CPU cores  : 4 @ 2445.404 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 7.8 GiB
Swap       : 4.0 GiB
Disk       : 55.1 GiB
Distro     : Ubuntu 22.04.3 LTS
Kernel     : 5.15.0-47-generic
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : UK Dedicated Servers Limited
ASN        : AS42831 UK Dedicated Servers Limited
Host       : 365 Online Technology Joint Stock Company
Location   : Coventry, England (ENG)
Country    : United Kingdom

fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 267.38 MB/s  (66.8k) | 3.69 GB/s    (57.7k)
Write      | 268.08 MB/s  (67.0k) | 3.71 GB/s    (58.0k)
Total      | 535.46 MB/s (133.8k) | 7.41 GB/s   (115.7k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 5.67 GB/s    (11.0k) | 6.07 GB/s     (5.9k)
Write      | 5.97 GB/s    (11.6k) | 6.48 GB/s     (6.3k)
Total      | 11.65 GB/s   (22.7k) | 12.55 GB/s   (12.2k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 1.01 Gbits/sec  | 1.77 Gbits/sec  | 3.80 ms
Scaleway        | Paris, FR (10G)           | 5.07 Gbits/sec  | 6.35 Gbits/sec  | 11.4 ms
NovoServe       | North Holland, NL (40G)   | busy            | 7.48 Gbits/sec  | 8.93 ms
Uztelecom       | Tashkent, UZ (10G)        | 2.02 Gbits/sec  | 1.79 Gbits/sec  | 88.1 ms
Clouvider       | NYC, NY, US (10G)         | 54.5 Mbits/sec  | 29.0 Mbits/sec  | 75.2 ms
Clouvider       | Dallas, TX, US (10G)      | 32.0 Mbits/sec  | 84.2 Mbits/sec  | 110 ms
Clouvider       | Los Angeles, CA, US (10G) | 30.8 Mbits/sec  | 59.8 Mbits/sec  | 128 ms

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 920 Mbits/sec   | 1.77 Gbits/sec  | 3.79 ms
Scaleway        | Paris, FR (10G)           | 7.46 Gbits/sec  | 6.09 Gbits/sec  | 11.3 ms
NovoServe       | North Holland, NL (40G)   | 9.22 Gbits/sec  | busy            | 8.93 ms
Uztelecom       | Tashkent, UZ (10G)        | 2.39 Gbits/sec  | 692 Mbits/sec   | 87.6 ms
Clouvider       | NYC, NY, US (10G)         | 51.8 Mbits/sec  | 21.9 Mbits/sec  | 73.2 ms
Clouvider       | Dallas, TX, US (10G)      | 35.6 Mbits/sec  | 81.8 Mbits/sec  | 109 ms
Clouvider       | Los Angeles, CA, US (10G) | 30.4 Mbits/sec  | 47.5 Mbits/sec  | 128 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 1496
Multi Core      | 4884
Full Test       | https://browser.geekbench.com/v6/cpu/2539714

YABS completed in 12 min 17 sec
  • MumblyMumbly Member

    @lowest, you should not be allowed to use the internet unsupervised.

    Thanked by 3bdl angstrom Arkas
  • angstromangstrom Moderator

    Thread closed on request of OP

This discussion has been closed.