-
Posts
10,125 -
Joined
-
Last visited
-
Days Won
545 -
Speed Test
My Results
Everything posted by CA3LE
-
ISP Test Log Average Enhancements
CA3LE replied to Pgoodwin1's topic in Ideas to make testmy.net better?
As always, thank you for your excellent suggestion. My plan is to basically duplicate what they do over at PassMark's site cpubenchmark.net https://www.cpubenchmark.net/compare/5027vs5684vs5172vs2830/AMD-Ryzen-9-7900X-vs-Intel-i9-14900KF-vs-AMD-Ryzen-5-7600-vs-Intel-Xeon-E5-2667-v4 Definitely won't be easy. But comparison like that is extremely helpful. Side note: pretty sure I'm going with the Ryzen 7900X on the new database server. Currently running Dual E5-2667 v4. The 7900X hits where I need it most (single thread) and a single CPU will blow the doors off what I'm running right now. ... couple that with Ryzen's ECC memory support and then they say that the AM5 socket will be the platform until 2027 so upgrade options for many years. -- when you run a mysql database query, that's a single thread operation. Excited to get it built and compare performance. So imagine that comparison but with internet benchmarks. Easy to understand and easy to use. -
Hello! How can I add a speed test server and company to the database? https://testmy.net/hoststats I don't have a way for you to be added to the database, it's automatic. Based on GeoIP data. Clients that test from your IP blocks may or may not be detected as Cube-Host depending on your IP leasing. If you own the IP space, then it should resolve to your name. I update my GeoIP data a few times a year. ... when TMN detects a client in your IP space it creates the databases, etc automatically. This can be checked by testing the source https://www.maxmind.com/en/geoip-demo and comparing to testmy.net IPTools If you'd like to host speed test servers, you can enable the beta (under My Settings) and then see https://testmy.net/myserver. You can use that to quickly setup test servers wherever you want. Or you can donate VPS servers to TestMy.net and I can build them out into full test servers. Let me know if you need any help, - Damon
-
Pgoodwin1 Hi CA3LE. Got an odd bug you can maybe fix. I go to DB/Fastest ISPs/Full Listing, I searched on Cincinnati Bell and Spectrum. I pulled up the Speed Test Logs for both - separate windows to compare It defaults to All Identifiers in the Results tables. I set it to 1000 / page. OK. I select the “Speed” column heading so that it shows the fastest. OK. Everything’s fine. The issue: In the Results Table, I can’t choose to display the fastest Speeds first in the Speed column when selecting an Identifier (other than All Identifiers) If I select an Identifier (like Dallas, TX), the Speed column automatically reverts to the slowest Speed of the 1000. When I select the Speed column heading to make it show fastest first, the Identifier automatically reverts to “All Identifiers”. When I reselect Dallas TX, it automatically displays the slowest Speed first again. haha - an endless cycle of despair Pgoodwin1 BTW: My personal test results “My Results” table doesn’t behave that way. You can select an Identifier then sort on the fastest speed without it reverting to All Identifiers Pgoodwin1 @CA3LE just wondered if you had a chance to look into this. If you don’t plan to, that’s OK too. CA3LE I'll fix this for you and hit you back. Pgoodwin1 😁 Great CA3LE Did some updates to the database query. It should now work the way you were expecting. Let me know if it's querying correctly now. Pgoodwin1 It’s working perfect now. Thanks. That’s a very cool tool. maybe in the future you could add the date range field like our personal results page has. That way you could look at the ISPs performance over a specific period. one suggestion on our date range in our personal results: make it so you don’t have to scroll through the Calendar. When you want the start date to be 3 years prior to today’s date, it takes a long time to get there. Maybe use two boxes, one for start date, one for end date, and let the user be able to type in the date as an option. More error prone that way though, but if it showed the date format, us idiots should be able to get it right. i don’t want much. HAHA🤗 CA3LE 14 hours ago, Pgoodwin1 said: maybe in the future The future is now. You can now search hosts and locations by date range. Currently this pulls data from a table where the old results are purged. So will be limited to about the last 45K results per host. Within those results you can now perform more fine grained searches. I'm working on a super server upgrade that will cut down query times on my larger tables. This should enable us to run queries like never before on the larger datasets. I have what I think is going to be a very interesting topic on the subject. Where I'm going to detail my build process. All of my servers are custom built but this new server is going to be my most custom build ever. Basically, found that enterprise grade hardware (that I've been using) isn't the way for TestMy.net moving forward. I was able to beat TestMy.net's current Dual Xeon E5-2667 v4 w/ Optane setup using a single Optane and an old i7 10700K. Not by a little either. It seems core performance is the bottleneck. The 13th and 14th gen Intel chips have oxidation and voltage issues so I think I'm going to end up going with AMD. Also Ryzen supports ECC memory. I'm basically building a Ryzen gaming PC inside of a 2U rack mount chassis. Also building a piKVM v2 into it to basically give the machine IPMI. Be able to remotely see the display, control input, mount ISOs... control it like I'm there. I've got the build coming in right now at less than $1K. That's about what I paid for just the CPU's and Optane drive in the master server. And I got them used, 1/6 the original price. After testing I'm convinced, my dream server is a gaming PC. Enterprise hardware, even the most recent CPUs... right there with basic consumer CPUs. They support more memory, have more cache but the performance per core where it counts most is far, far lower. My concern with hosting TestMy.net is usually focused on single thread performance. Similar to how I focus on single thread performance in my tests but this is talking about CPU threads. More threads is definitely helpful and nessesary but the speed of each thread is the most important variable. The single thread performance I'll get out of any AM5 Ryzen will absolutely crush any of the Xeon CPUs in my price range. The ones I'm looking at out perform or are at par with even the latest enterprise CPUs. At the end of the day I can build servers cheaper with the latest chips. All working together in my proxmox cluster so hardware failure is tolerable... but I imagine they'll be just as stable, under volting the CPUs and run them through extra stability testing before production. There's so much competition in the gaming PC performance arena which leads to less expensive consumer chips being faster. And yes, better database structure solves problems. But so does adding more power. It's far easier for me to give my cluster more power than to restructure databases. Things I've built recently and going forward are better planned but there's all the databases and tables prior. If you don't mind, I'd like to share this thread as a topic. Pgoodwin1 Wow. That sounds really good. It does totally amaze me what you can get in processing power these days. Consumer grade machines can be so fast and not that expensive. “If you don't mind, I'd like to share this thread as a topic.” Certainly go ahead and share it as a topic thanks for the upgrade with the date range. That tool is awesome now for comparing ISPs. I’ve been looking at maybe changing from Spectrum to AltaFiber (which is Cincinnati Bell). That change you made just adding the date range makes it very easy to compare apples to apples - different test servers, No problem, thank you for the suggestion. CA3LE No problem, thank you for the suggestion. The limitation on the host and location queries has needed that correction for a long time. Once I know someone is trying to query in a broken way, it's motivation to make it work the way you expect.
-
When the auto speed test was first designed in 2010/2011, phones were still figuring things out. Always, my goal with design is to make ONE thing that works with all browsers and devices. So I've always avoided apps. TestMy.net has never had an Android of iPhone app. Now it's 2024, phones are a different breed. They are full fledged gaming PCs that make the desktop computers from 2010 look like a joke. In 2010 average RAM was 3-4GB. Now, our phones are coming out with 8-12GB! The philosophy remains, ONE thing for them all. For the auto test to work on any machine, not just Android, make sure the power remains on and the test in focus. On my S22 I normally have my display set to never turn off. I'm now noticing that they must have recently updated... now you can't set anything higher than 10 minutes. That's alright, there's an app for that. Screen Alive on Android does the trick. It's free with an option to donate to the developer. Surprisingly, this isn't an issue on iOS. I would have thought if anyone did this it would be Apple. Android usually does things first but not things like this. On iOS just go do Display & Brightness > Auto Lock > Never. Once your phone's display doesn't sleep anymore you can plug it into power and start up the auto test. Keep the browser in focus and it will do its thing. Please let me know if this helps
-
It may have been a routing issue. Has it cleared up yet? If I saw something like that happen as @xs1 said, I'd run traceroute and compare to a known working machine. As for getting it fixed. Depends on where the issue is, could be on your provider's end, my provider's end or one of their peers in between. Usually, these kinds of problems resolve on their own. It may be that there was only an issue going to and from that specific datacenter. As long as you're testing good across the greater majority, I wouldn't worry about it too much. Unless it starts affecting the real communication that you do. Important thing... we know it's not unique or localized to your computer or network. You've proven this by testing across multiple networks and showing greater results.
-
Confused about average results - how about a histogram?
CA3LE replied to wmertens's topic in Ideas to make testmy.net better?
The stats that show under hoststats are taken over longer periods and have more results considered. The stats that show under the ISP ranking tab are done in the background every 15 minutes and only consider the most recent results. My next major hoststats update will use the same numbers for everything. Using the overall average shown on the current hoststats pages. I think a histogram is an excellent idea, I'll be working on that. With less than 300 results each, it's hard to compare. The dataset on those providers is very small. If you click the "Log" button on the hoststats pages you can see the number of results. iWay looks better... but Init7 has a slightly larger sample. -
Generate upload data in-browser
CA3LE replied to wmertens's topic in Ideas to make testmy.net better?
Sorry it's taken me a couple of days, I'm developing. If you toggle the beta in My Settings, then visit the upload or download test there's an explanation. There are reasons I originally did it that way and they still hold true. It makes for a more consistent test with far less variables. Man! Your suggestion for the upload test back in 2018. That's an algorithm I still use. Has saved a lot of bandwidth and time... no joke, over the hundreds of millions of upload tests since, probably something like a decade or more of wait time saved at this point. I can calculate the bandwidth saved... I'll just estimate quickly, going off the recent upload test results Keep in mind, what's logged to the database is the final result. The client may have cycled though up to 4 tests before getting to the final test. 2 GB in the last 4 minutes (non-peak early morning hours) 1440min per day / 4minutes = 360 * 2GB = 720GB per day The optimization was made 6 years ago 720*365*6 = 1,576,800 GB or 1.6 Petabytes! Wow. I think you could easily figure an extra 30+% for the pre-tests. So about 2 PB saved! Using the current median download speed from the recent download test results of 62 Mbps we can get a rough estimate of the time saved. byte conversion 62/8 = 7.75 MB/s 2 PB is 2147483648 MB (2*1024*1024*1024) 2147483648 / 7.75 = 277094664 seconds saved (((277094664 / 60sec per min) / 60min per hr) / 24hrs per day) / 365days per yr = 8.8 YEARS SAVED! The actual number is probably much higher. Your post, I'd say has saved at least 10 years of wait time collectively. Amazing. I'm keeping your histogram idea at the forefront of my mind too. -- I'll post over there. -
Hi Z! What it's actually looking for by default is the presence of favicon.ico. Your website doesn't have that in the document root. Just now I made it work the way that you were expecting. If favicon.ico isn't found it then checks for /b/img/blank.png. blank.png is preferred. Hope this helps. Here's my result https://testmy.net/latency?tID=g5uvfbbcq3
-
If you see the reference value, you're well within the spec. Between -27 and -8 dBm Those numbers look good to me.
-
Welcome to TestMy.net @James2007 Let me know if you have any questions or comments.
-
Hi Sean, I made some adjustments, when you enable "Linear Load" it will modify the Upload page to direct you to the legacy upload test. This doesn't work with combined test, using combined test with linear load under the beta will still load the new version. I hope to develop the linear upload test to work with the new graphs but I'm not sure if it's possible yet. It's pretty tricky. Porting the download test over was simple. The problem has to do with a limitation of a certain browser API. The current version will remain available, I have no plans to get rid of that. If I can't get it to work exactly the way I want... then I'll just integrate the old version, minus the extra graphing on the upload.
-
Reading the speed between their server and your computer is not really testing the internet speed. Doesn't matter how you connect. You're not testing the Internet if you don't actually go out to the Internet. Testing against your ISP's servers isn't going out to the Internet. That's why TestMy.net is here. Your ISP has control over the quality of peering and bandwidth in and out of their network. If your ISP is in the UK, then it should be a very quick hop over to my UK servers. As long as your ISP is delivering that won't affect the final result. There would be little to no difference if all the connections between are running with capacity available. Any good ISP in that scenario would have at least multiple 10 GbE peers, meaning that the route between the ISP and TestMy.net should never be the weakest link... unless it's over capacity. It's a red flag if an ISP tells you that only testing against their server's is accurate. And like I said, it doesn't make sense in the first place. Your ISP's servers are not the Internet, that's your host's network... before the Internet. It's a part of the Internet but if that's your host, their network is your network. A step above your local area network but we're not really out to the wider Internet. TestMy.net is actually testing your connection out to the Internet.
-
Hi Steven, welcome! When you test at TestMy.net you're testing the connection from your home, through your providers network, out to the internet and then to my servers. You only need to consider your own location when choosing one of my locations to test from. Usually TMN will do a good job of choosing for you. Based on your IP address, TMN would pick UK servers to test from. You can also visit the Mirror page and quickly test your latency across all of the locations. UK will again most likely perform the best (lowest). After you're settled on a test server location then head over to the Auto Speed Test to schedule automatic testing. Once you've gathered some results (data) visit My Results and My Average to get a better understanding of the results. Hope this helps, please let me know if you have any other questions.
-
The beta is now available to all members. You'll find a toggle switch in My Settings. Hope you find this helpful.
-
Here's a photo I took yesterday. We had 77% coverage where I'm at. A Sun spot made it look kinda like Pac Man for a little bit. (the sun spot was visible in other photos so I know it wasn't an artifact) Taken on an S22 Ultra, using the 10X lens through solar filter glasses, manual focus and exposure.
-
You're already a beta tester, visit the beta introduction to get started.
-
Welcome @dudxs! Check out https://testmy.net/database. I store results based on GEO location and ISP.
-
First, All of the the database results are from TestMy.net only. There are also several ways you can programmatically extract the averages. I'll use Comcast as an example but this works for locations as well. Basic json of what you see rendered on Comcast's hoststats page https://testmy.net/hoststats/comcast_cable?jsonout=1 CSV output to file https://testmy.net/hoststats/comcast_cable?csvout=1 CSV output rendered in page https://testmy.net/hoststats/comcast_cable?csvout=1&easycsv=1 You can also specify how far back and output the averages https://testmy.net/hoststats/comcast_cable?csvout=2&monthsback=12&easycsv=1 Note for the output, I use pipes (e.g. " | ") as a separator. Below I'm adding line breaks to make it easier to read. "April 2022","May 2022","June 2022","July 2022","August 2022","September 2022","October 2022","November 2022","December 2022","January 2023","February 2023","March 2023","April 2023","May 2023","June 2023","July 2023","August 2023","September 2023","October 2023","November 2023","December 2023","January 2024","February 2024","March 2024" |219.5,192.5,190.6,156.1,157,210.8,165.5,224,215.5,215.5,209.6,168.4,206.6,203.6,220.9,270.5,272.4,218.3,244.9,240.8,299.5,249.9,312.7,190.5 |9.9,17.4,15.2,16,14.6,17.7,37.1,16.5,13.6,15.6,17.7,17.1,18.6,14.2,21.3,22.1,22.9,24.2,27.8,34.1,28.1,25.7,26.7,22.9 |'NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL','NULL',95,92,89,92,123,107,91,114,104,75,119,81,127,79,78,106,82,76,97,89,95,194,71,61 Translates to... Date Field |Download |Upload |Latency You can make a simple program to CURL and parse that output, explode and then make arrays out of the data. Make sure you set a reasonable limit to how often you query or it will be seen as a threat. And please share with others where you're getting your data. TMN is 100% word of mouth.
-
Was that only happening on the My Results page or did you see that on the test results as soon as it finished the test also? I'm loading a different version of the charts for you on My Results, it should be working normally there. Your helping out great. Keep in touch with me and we'll get this bug nailed down. I appreciate the help.
-
That's a bug with the charting program. I thought I resolved this. I'll try to duplicate the issue on my end so I can nail it down finally. Odd because I can't replicate it this time... at least not so far. I'll update this topic to have you test again after I make a few edits.
-
See if you can do https://gifcap.dev/ and capture what's happening. ... or search "screen record chromeos" --- looks like they made it easy on there.
-
Ran ChromeOS, tested the beta and didn't see the behavior you described. Actually that particular computer ran better under ChromeOS than MacOS... I might have to install it full time on that machine. It's a 2011 and has been progressively slowed down by Mac updates. Funny, all the problems seem gone, same exact hardware different OS. Proves that it's all software and personally, it feels intentional. Anyway, I tried to install it first on my Virtual Machine setup using proxmox but the RAW file makes it a little annoying. I need an ISO. So I opted to just put it on a flash and run it on my hardware test bench. Unfortunately I was met with a stuck boot screen. It showed the ChromeOS Flex splash screen but wouldn't load further. That's when I turned to one of my old iMacs. Booted right up and tested as I would expect (running on the wifi under suboptimal conditions but that's not what we're testing). Do you have any Chrome extensions installed? Go to chrome://extensions/ in your browser, let us know what you have installed. Post a screenshot. Basic run down of what can happen sometimes... Extensions that parse (read) the data in your browser may be overwhelmed by the data TMN dishes out. This can cause weird things to happen. If you see an extension that's doing this, e.g. you disable an extension and the symptom disappears, then chances are... that extension is parsing the data in your browser. In other words, if it can do that... it can spy on you as well. I would not use an extension with that sort of behavior. It may be "legitimate" because it's a "deal catcher" extension or something like that... it may need to spy on you to get you that deal. And you're giving it that permission. To hunt down the extension, systematically disable extensions and then re-test until you nail it. If you have a bunch installed, maybe start with just disabling all of them. Then re-enable the ones you really need. The current version passively detects things like this all the time. Something special about my process... other speed tests don't seem to do this. According to our users that run into these kinds of extension issues. I just want to compare our user agent strings to be sure... Yours (taken from your test results): Mozilla/5.0 X11; CrOS x86_64 14541.0.0 AppleWebKit/537.36 KHTML, like Gecko Chrome/113.0.0.0 Safari/537.36 Mine: Mozilla/5.0 X11; CrOS x86_64 14541.0.0 AppleWebKit/537.36 KHTML, like Gecko Chrome/113.0.0.0 Safari/537.36 ... exact match. So there has to be another variable. Cases like this, it's most often extensions. If you really need that "deal catcher" extension (or whatever it is) I suggest disabling it, until you actually need to use it. Once you're done, turn it back off. There are very few exceptions where I feel an extension is necessary. Usually, you give up way more permissions that you need to, they are a major security risk. I don't use them myself and I recommend my users also steer clear.
-
I develop primarily using Chrome so I'm not sure why that would be happening. But maybe there's something different about ChomeOS. I'm installing ChromeOS for testing and will report back on this topic.