Zores
join:2002-02-24
New Milford, NJ Reviews:
 ·Optimum Online
| So I hopped on the 150 mbps train and I get erratic results My best being this
»speedtest.net/result/1735791302.png
My worst being this
»www.speedtest.net/result/1735863801.png
I checked the ONT and the Ethernet port is flashing so that means I should be on 1000 mbit. Also checked the Actiontec Rev. G and WAN Ethernet is on.
What should I do? |
|
|
|
 MrKal_El
join:2003-01-19
Franklin Square, NY | I was getting the same results earlier this week and after calling up FiOS and having them re-provision me I am getting much much more consistent results.... (Also, a reset of the router doesn't hurt)
--
Twitter: MrKal_El |
|
Zores
join:2002-02-24
New Milford, NJ | Are you using the default actiontec? |
|
 tfrionliTom F.
join:2001-06-21
Kings Park, NY
 ·Verizon FiOS
·Optimum Online
| reply to Zores
Use the Verizon speed test appropriate for your area and speed.
»www22.verizon.com/residentialhel···5271.htm
Then optimize your connection, make sure your are doing a test hardwired to avoid any wireless issues.
--
tfrionli |
|
| reply to Zores
I noticed both of the tests use the same server.
When I get unusually low results I try another server and it always has been a problem with the server so far. |
|
| reply to Zores
With such high speeds the problem could very well be the server your testing. Many can't handle 150 Mbps if they are under or close to load.
I suggest you test during off-peak hours (like at 1:30 AM which is when you got your best results) over the next few days. If you consistently get what your paying for during off-peak hours the problem lies with the server your testing and not your connection. |
|
Zores
join:2002-02-24
New Milford, NJ Reviews:
·Optimum Online
1 edit | reply to Zores
I'm starting to think something is a little off.
I rarely hit 150 mbits/s and on my previous 35/35 connection I would get the same speeds always. I don't think FIOS at 150 mbps would be that saturated here so it feels like something is off.
Broadband Speed Test
Analysis information:
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
SendBufferSize set to [261360]
running 10s outbound test (client to server) . . . . . 64.07Mb/s
running 10s inbound test (server to client) . . . . . . 135.88Mb/s
------ Client System Details ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Sun Microsystems Inc., Version = 1.6.0_29
------ Web100 Detailed Analysis ------
Client Receive Window detected at 261360 bytes.
10 Gbps 10 GigEthernet/OC-192 link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.
Web100 reports the Round trip time = 12.7 msec; the Packet size = 1452 Bytes; and
No packet loss - but packets arrived out-of-order 0.03% of the time
This connection is receiver limited 2.57% of the time.
Increasing the the client's receive buffer (255.0 KB) will improve performance
This connection is sender limited 83.66% of the time.
This connection is network limited 13.76% of the time.
Web100 reports TCP negotiated the optional Performance Settings to:
RFC 2018 Selective Acknowledgment: ON
RFC 896 Nagle Algorithm: ON
RFC 3168 Explicit Congestion Notification: OFF
RFC 1323 Time Stamping: OFF
RFC 1323 Window Scaling: ON
Information: Network Middlebox is modifying MSS variable
Server IP addresses are preserved End-to-End
Information: Network Address Translation (NAT) box is modifying the Client's IP address
Server says [108.50.219.211] but Client says [192.168.1.2]
EDIT: Maybe I'll call them up and try what you said MrKal El |
|
MrKal_El
join:2003-01-19
Franklin Square, NY | reply to Zores
said by Zores:Are you using the default actiontec?
Yup...
--
Twitter: MrKal_El |
|
MrKal_El
join:2003-01-19
Franklin Square, NY | reply to Zores
said by Zores:I'm starting to think something is a little off.
I rarely hit 150 mbits/s and on my previous 35/35 connection I would get the same speeds always. I don't think FIOS at 150 mbps would be that saturated here so it feels like something is off.
Broadband Speed Test
Analysis information:
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
SendBufferSize set to [261360]
running 10s outbound test (client to server) . . . . . 64.07Mb/s
running 10s inbound test (server to client) . . . . . . 135.88Mb/s
------ Client System Details ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Sun Microsystems Inc., Version = 1.6.0_29
------ Web100 Detailed Analysis ------
Client Receive Window detected at 261360 bytes.
10 Gbps 10 GigEthernet/OC-192 link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.
Web100 reports the Round trip time = 12.7 msec; the Packet size = 1452 Bytes; and
No packet loss - but packets arrived out-of-order 0.03% of the time
This connection is receiver limited 2.57% of the time.
Increasing the the client's receive buffer (255.0 KB) will improve performance
This connection is sender limited 83.66% of the time.
This connection is network limited 13.76% of the time.
Web100 reports TCP negotiated the optional Performance Settings to:
RFC 2018 Selective Acknowledgment: ON
RFC 896 Nagle Algorithm: ON
RFC 3168 Explicit Congestion Notification: OFF
RFC 1323 Time Stamping: OFF
RFC 1323 Window Scaling: ON
Information: Network Middlebox is modifying MSS variable
Server IP addresses are preserved End-to-End
Information: Network Address Translation (NAT) box is modifying the Client's IP address
Server says [108.50.219.211] but Client says [192.168.1.2]
EDIT: Maybe I'll call them up and try what you said MrKal El
Let me know how it goes...really only take a 10 min phone call...
--
Twitter: MrKal_El |
|
 RoKmAn
join:2000-09-09
East Elmhurst, NY | Speed tests are NOT a reliable way to gauge the performance of this tier. Most speed tests servers are not capable of consistently pleasing this sort of speed. Use a download accelerator or download bunch of stuff from different sites and use a bandwidth meter to see whats up!
35/35 would show reflect the speeds consistently is due to the speed tests servers consistently having such a capacity. |
|
 wmcbrine213 251 145 96
join:2002-12-30
Laurel, MD | reply to Zores
I like how it falls off thirty-fold, yet that still only takes it down to "Grade: B (faster than 72% of U.S.)". 
--
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0 |
|
Zores
join:2002-02-24
New Milford, NJ Reviews:
·Optimum Online
| reply to RoKmAn
This might be true of the downstream but even upstream?
On my 35/35 package my upstream was always a steady 32 mbits at Fortress ITX Clifton Speedtest.net. Has been steady from 2009-2011.
Now on the 150 mbits package my upstream sways from 10 mbits to 50 mbits on the same speedtest.net server. Can you explain this? |
|
 Smith6612Premium,MVM
join:2008-02-01
North Tonawanda, NY
kudos:21 | reply to wmcbrine
Goes to show all of the crappy connections running out there, lol. 1Mbps is an F on Speedtest.net |
|
Reviews:
·Verizon FiOS
·Charter
·AT&T U-Verse
 ·DSL EXTREME
| reply to Zores
said by Zores:SendBufferSize set to [261360]
Increasing the the client's receive buffer (255.0 KB) will improve performance
Your receive window is set WAY too low. That would really even be a little low for 35/35. Try upping that by 4-5x that amount.
--
150/75 mbit Verizon FiOS connection FTW! |
|
htin11
join:2000-08-10
Flushing, NY | reply to Zores
i don't think you should count on the verizon speed test. I have 150/35 too and i never get that on verizon speed test. However i get 153/56 on www.speedtest.net going thru a datacenter in NJ. |
|
Zores
join:2002-02-24
New Milford, NJ Reviews:
·Optimum Online
| reply to houkouonchi
Good call houkouonchi, u rock!
Yay, that fixed the downstream at least. Downstream seems rock solid now just like before on 35/35, always hitting 153 mbps at Fortress ITX, Clifton, NJ.
Now, how do I fix my upstream? You'll see in the screenshot it is not steady at all. On 35/35, it was always rock solid too.
Edit: I used FIOS Optimizer with IE and admin priviledges and it gave me erratic downstream. Used TCPoptimizer from speedguide.net and now my downstream is fixed. |
|
Reviews:
·Verizon FiOS
·Charter
·AT&T U-Verse
·DSL EXTREME
| said by Zores:Good call houkouonchi, u rock!
Yay, that fixed the downstream at least. Downstream seems rock solid now just like before on 35/35, always hitting 153 mbps at Fortress ITX, Clifton, NJ.
Now, how do I fix my upstream? You'll see in the screenshot it is not steady at all. On 35/35, it was always rock solid too.
Edit: I used FIOS Optimizer with IE and admin priviledges and it gave me erratic downstream. Used TCPoptimizer from speedguide.net and now my downstream is fixed.
NP! Yeah 150 megabit can really require some extreme values because when ur doing 18+ megabytes/sec a measily 256 kilobyte buffer gets filled in milliseconds =)
As for your upstream this is one thing that speedtest.net really does badly at. Honestly once you get over 70-80 megabits outbound it stops being able to keep up because of the way it tests. Even from a 1000/1000 I can get > 300-400 megabits on tons of speedtest.net servers downstream but only 1 or 2 can give me over 60-70 up. Also I have noticed even when it gives me that result it never fully utilizes my upstream on my netgraphs (even for my home connection which usually yeilds about mid 60s up on speedtest.net (and 72+ other methods). Even when I bonded two 35/35 connections the upload was too high for speedtest.net I would get 68 down and never above 60 up even though testing to the same server using ndt yeilded 66/66 (along with Vz's test). Part of the lowered results was the overhead from my VPN bonding (instead of being 70/70).
A good way to test upload is via FTP or something else and use something like netmeter to see what you get or try seeding a bunch of torrents with a lot of leechers =). I do find VZ's test is usually pretty good on that though and your result is a little low as well (at 64 as most are closer to 70-74). I also seem to recall people mentioning that NJ gives bad upstream results, what does the NY server give you? Any difference in stability/results? Your send buffer could also use a boost as well but not convinced that is your entire issue especially as if anything that might cap your speed on a single TCP/IP transfer it really shouldn't make the results inconsistent.
--
150/75 mbit Verizon FiOS connection FTW! |
|
Zores
join:2002-02-24
New Milford, NJ Reviews:
·Optimum Online
| Is there a way to set the numerical values manually in Windows 7 x64 like there was on XP x86? I Googled but all I could find was TCPOptimizer which doesn't really have anything you can set like that. I remember fiddling with my RWIN values a lot on DrTCP a few years back. I know that Windows 7 RWin was supposed to auto scale (but doesn't do it well enough it seems) so I never messed with it this time around.
As for the upload it is still weird. You can see in my screenshots the value hovers to fortressITX. But when I had 35/35, it always stayed around 32 mbits at all times of day. It's why I suspect something is up.
EDIT: All the NY speedtest give me even worse results if you can believe that.
DOUBLE EDIT: This is what I get now after running TCPOptimizer.
Broadband Speed Test
Analysis information:
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
SendBufferSize set to [261360]
running 10s outbound test (client to server) . . . . . 64.29Mb/s
running 10s inbound test (server to client) . . . . . . 148.25Mb/s
------ Client System Details ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Sun Microsystems Inc., Version = 1.6.0_29
------ Web100 Detailed Analysis ------
Client Receive Window detected at 554496 bytes.
10 Gbps 10 GigEthernet/OC-192 link found.
Link set to Full Duplex mode
Information: throughput is limited by other network traffic.
Good network cable(s) found
Normal duplex operation found.
Web100 reports the Round trip time = 19.72 msec; the Packet size = 1452 Bytes; and
No packet loss - but packets arrived out-of-order 0.01% of the time
This connection is receiver limited 2.7% of the time.
Increasing the the client's receive buffer (541.0 KB) will improve performance
This connection is sender limited 92.41% of the time.
This connection is network limited 4.9% of the time.
Web100 reports TCP negotiated the optional Performance Settings to:
RFC 2018 Selective Acknowledgment: ON
RFC 896 Nagle Algorithm: ON
RFC 3168 Explicit Congestion Notification: OFF
RFC 1323 Time Stamping: OFF
RFC 1323 Window Scaling: ON
Information: Network Middlebox is modifying MSS variable
Server IP addresses are preserved End-to-End
Information: Network Address Translation (NAT) box is modifying the Client's IP address
Server says [108.50.219.211] but Client says [192.168.1.2] |
|
Zores
join:2002-02-24
New Milford, NJ Reviews:
·Optimum Online
2 edits | reply to houkouonchi
Finally found a stable test for upload. Testing to Comcast and I'm stable at 35 mbps and above for upload. I guess FortressITX is having issues or the routing for 150 mbps here in my new place is different than 35 mbps my old place. *shrug*
Another test location. Hmmm, lol.
|
|
| reply to houkouonchi
said by houkouonchi:said by Zores:SendBufferSize set to [261360]
Increasing the the client's receive buffer (255.0 KB) will improve performance
Your receive window is set WAY too low. That would really even be a little low for 35/35. Try upping that by 4-5x that amount. I am having the same problems, I have Win7 How do i increase my sendbuffersize to [261360] ?
Analysis information:
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
SendBufferSize set to [261360]
running 10s outbound test (client to server) . . . . . 74.63Mb/s
running 10s inbound test (server to client) . . . . . . 118.99Mb/s
------ Client System Details ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Sun Microsystems Inc., Version = 1.6.0_30
------ Web100 Detailed Analysis ------
Client Receive Window detected at 65536 bytes.
100 Mbps FastEthernet link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.
Web100 reports the Round trip time = 4.23 msec; the Packet size = 1460 Bytes; and
No packet loss - but packets arrived out-of-order 0.15% of the time
This connection is receiver limited 73.52% of the time.
This connection is sender limited 26.08% of the time.
Web100 reports TCP negotiated the optional Performance Settings to:
RFC 2018 Selective Acknowledgment: ON
RFC 896 Nagle Algorithm: ON
RFC 3168 Explicit Congestion Notification: OFF
RFC 1323 Time Stamping: OFF
RFC 1323 Window Scaling: ON
Packet size is preserved End-to-End
Server IP addresses are preserved End-to-End
Information: Network Address Translation (NAT) box is modifying the Client's IP address
Server says [108.46.143.117] but Client says [192.168.1.4] |
|
