Pgoodwin1

How is that percent calculated? I'm talking about when you're in your results list, and you pull up a download TID plot, and it tells you what the variation is. It seems like it's giving bigger percentages than it used to.

amazing. i can almost see the "aura of miracle" about you. i used to ask the other engineers and techs in our engine control electronics lab where we breadboarded circuits if they could see the aura of miracle about me.

Haha. Good that it was real and you found it quickly - fixed in less than 2 hours is pretty awesome. Thanks

after running a Combined test (which runs fine), I hit the Re-Test button. the test appears to start but very quickly (10th of a second or less) it aborts to this page: https://testmy.net/download it started doing this only very recently. it's always run OK in the past.

Any site that gives you results significantly higher than your plan 125 down / 12 up is highly suspect. I wouldn't trust the reading to be giving you anything representative of what your system is doing. The sites that are giving you less than 100 Mbps are probably telling you some version of the truth. Since testmy.net is one of them, believe it. It's highly like that during the hours of 10 AM and midnight, your speeds may vary because of traffic outside of testmy.net. If you take a reading at like 2 AM on this site! you'll probably get a reading that is the truth and should be very close to your 125/12 plan. That reading may drop down during daytime hours. Test with an Ethernet cable connected. Then check if your wireless speeds is approaching the Ethernet performance (assuming you are using wireless).

Some of the issue is the distance. But the floors and walls and what's in them make a difference too. Also, be sure there's a good amount of space around the wireless router. When I'm out in the garage, it's about 30 ft, and it's through one wall and part of a pantry. I get 1/3 to 1/2 my wired speed, and sometimes less than that. You can usually look at your signal strength icon and look at how many bars you have. If it's less than max indicated, your speeds will likely be quite a bit lower than the wired speed - sometimes way less, like your 1/8 or less. But even at long distances, when the signal strength icon is at max the speed will likely be considerably lower than your wired speed.

Speedtest overstates your performance. Go to TestMy.net Home page and read the tab " What Makes TMN Different?"

Strange that they decided to leave the shares named GOOG.

see if your speeds here on TMN are slower during the same hours. mine typically are slower by 10-20% between dinner time and 11 PM or so. it's a little worse than it was a year ago when i got on the 50 mbps down and 5 mbps up plan at Time Warner. i've noticed that it's still pretty fast in the morning up until 11 AM or so. from 11 AM to 11 PM it's spotty. one minute it's full speed, the next minute it's way down (sometimes to 60-70% of max. there must be a lot of people streaming video

TMN does have a mobile option. Go to the bottom of the browser window and there's a link to it. My biggest variation in download speed on my mobile devices is due to the signal strength of the wireless where I'm physically at in the house. That plus the speed of the device itself. My iPhone 5s gives me slightly higher speeds than my iPad 2.

If you want your users to get an accurate answer, put a link to TestMy.net on your site. Since the owner of TestMy has devoted 19+ years to developing a system to measure download and upload speeds! I doubt that he'll be willing to just hand you the capability.

I power cycled the D-Link router without saving the event log, so I don't have anything to look at from a month ago. The cable modem status event log didn't erase itself and showed a couple of "Lost MDD Timeout" events and a "No Ranging Response Received" events about a month ago, but I have no idea what those mean.

I noticed about a month ago that my speeds via Ethernet to TIme Warner Cable had dropped. I power cycled the cable modem and my router today and got my speeds back to where they were. The trend plot was very helpful CA3LE. I guess every once in a while you need to clear their brains out. Not sure if the Cable Modem or the Router did the trick but I'm guessing the modem. I was just about ready to call Time Warner on the lower speeds, but today I was by myself here and nobody using the equipment except me, so I power cycled everything.

a little while later, about 11 PM, the Multi and Single thread tests were about the same.

Well I tried Firefox vs Safari using Multithreading tonight. They both had very similar results. Both were a little slower than the single thread tests by about 20% maybe. I did the tests during peak hours - about 10:30 PM. My single thread results were only about 35 down where they are usually at 50 during off-peak hours. The Multithread tests were in the 28 Mbps down range. My plan has a 5 Mbps upload max which I was reaching during Combined testing tonight. Is it possible that the 5Mbps up limit comes into play more during Multithread testing and slows the process down? My mid 2010 iMac is only an i3 processor, but with nothing else running I wouldn't think that it would be the bottleneck. But I don't pretend to understand the Multithread process and how much it taxes the main processor.

Thanks for the responses. Good to know CA3LE: "When it's the other way around and your multithread results are MUCH faster... then it's more cause for concern." I'll try another browser on the iMac

Pretty much all of the hardware in the iPhone 6 having to do with speed is 1 or 2 generations advanced over the 4th gen iPad. Assuming your AirPort Extreme is a newer ac model, with the data processing and wifi input/output processing capability the phone has, when you test your phone while in close proximity to your AEBS, you may be hitting your ISP plan max speed. I read another article that said that their device was twice as fast an 802.11ac as it was at the same distance on 802.11n. So that, combined with the improved processors and memory speeds of the IP6, it's not surprising that there's a significant difference. However, I get about 37-40 Mbps on my iPad 2 with an older "n" AEBS, but only when the iPad is sitting right next to the base station. I get 45-53on my 5S (virtually the same as my iMac over Ethernet, When I move 30ft away and through a wall and some kitchen cabinets, it's only a little over 20 Mbps on the iPad.. So there's a lot of factors. Again, my ISP plan max is 50 down.

I saw one review of the 4g iPad and they got 37 Mbps. Another review of the iPhone 6 got 103 Mbps. Pretty close to what you got. So I'd say it's real. What's your plan max?

Well. The A8 processor in the 6 Plus can be "up to" 2.5x as fast as the A6X in the 4th gen iPad. The 6 Plus also has ac WiFi vs n for the iPad. Not sure if your wifi router is ac capable, but if it is, the ac speeds can be 3x the n speeds. The 6 Plus is a 64 bit machine where the iPad 4Th gen is a 32 bit. I don't know whether Apple uses a 64 bit data path between the processor and the WiFi I/O though. It does seem like a big difference though. My 5S is way faster at everything than my iPad 2 and like yours, the 5S processor is two generations newer. But both my machines run wireless right up to my 50 Mbps plan limit, so I can't compare like you're doing.

Try clearing all cookies and reset Safari. Most of the fixes I just looked at involved some cookie causing a conflict.

Pretty impressive. I'm not sure the Ethernet cabling in my house would even do those speeds. I think all my cables are only cat5.

After reading a lot of reviews a couple of years ago, the SB6141 was the one that I had chosen to buy. I instead upgraded my plan with TWC and they gave me the ARRIS as part of their Signature Home plan. It was better than the older DOCISS 2 modem I was using prior to that. It has never given me any problems so I chose not to buy the SB6141, but it was on the list of modems that TWC supported.

Thanks for the response. I don't have any 3rd party protection software, only what is built into Mac OS 10.9.5. That and whatever is built into my old D-Link DI-604 router. When I set it up, I didn't do anything special. Not sure if Time Warner is doing anything different during the Multithread testing. For single thread testing I almost always get slightly above their advertised 50/5 plan speeds. I'm not sure the slower multithread testing results are telling me anything about my performance in real world web page loading. Whenever the single thread test results are at or near the 50/5 maxes, I get good web surfing response except for site that I know are always slow.

Mine are always about 25-40% slower. From the description of Mutithread test here on TMN: "Enabling the multithread speed testing option on your download speed test can really open up your connection for maximum throughput." It doesn't matter what time of day I choose. Typically I would select the Wash DC and Dallas servers. When I add the west coast servers, it's even slower.

I see differences similar to what you're seeing here on my Arris TG862G cable modem connected to TWC on a 50/5 plan. I see the differences when I switch from a single thread test on the Washington server to the Dallas server. I always assumed it could be explained by instantaneous loads at TWC and/or whoever they rent Internet bandwidth from. I don't have the equipment or software to measure anything local. Seems like the only way to get a definitive answer on your two modems would be to disconnect them from the Internet and load them with a simulated Internet interface. I also don't really have detailed knowledge on any of this. All I have is a lot of data points