cancel
Showing results for 
Search instead for 
Did you mean: 

Rogers Hitron CODA-4582 Hardware

Double_K
I'm a Reliable Contributor
2,050 REPLIES 2,050

Re: Rogers Hitron CODA-4582 Hardware

arnym21
I Plan to Stick Around

@Datalink

 

Actually, NAT Acceleration switched off in a router would result at much higher speed loss than 100Mbps at Gigabit speeds, unless speed was originally quite low. What people forget, speed is not limited only by one's modem HW and FW, and neither by one's router. Many report that with rare exceptions, on consumer Windows PCs with CPU earlier gen than I5 and I7, wired speeds generally won't go higher 500-600 Mbps regardless of anything. And these numbers signify deeply optimized TCP stack, latest OS and Ethernet chipset drivers, all card packet processing offloads set On, carefully cleaned up system with minimal antimalware and traffic monitoring. Out of the "used box", on a Gigabit sub an older non-optimized system may easily max out at 150Mbps regardless of any modem and router choices, while most subscribers would probably still blame Rogers network "congestion" for slow speeds. Smiley Happy

 

Does anyone know what is exact config of Rogers Techs laptops, and what optimizations they run on them? I seen very impressive numbers these laptops can crunch on a DSLR speed test on a seemingly near clogged 1G sub as per subscriber equipment tests. Some commented, they bypass modem config speed restrictions either using special network login.

 
 
 

Re: Rogers Hitron CODA-4582 Hardware

Keep in mind that @Prospect might not be running the gigabit plan.  The 4582 can be used by anyone on an unlimited internet plan.  So, with that in mind, its possible to run a slower data rate without Cut Through Forwarding or Flow Acceleration enabled in the router and stay fairly close to the original data rate that you see out of the modem itself.  It depends of course on the number and type of functions running in the router.  @Prospect would have to provide additional details on his or her internet plan and the type of functions that are currently enabled in the R7000 in order for me to take an educated guess as to what the problem might be. 



Re: Rogers Hitron CODA-4582 Hardware

HughR
I Plan to Stick Around

@DatalinkI'm thinking about your NAT comment.

Network performance is really tricky.

My router is a little PC running CentOS 7 Linux.  The CPU is a Celeron 2961Y at 1.1GHz.  I don't imagine that it is the bottleneck, but I'm not sure how to tell.  There certainly is no NAT acceleration hardware.

 

Half an hour ago I got 910 from behind the router.

A minute ago I got 842 on the router and then 804 behind it.

Repeating: 852 on the router and then 876 behind it.

Repeating: 851 and then 785 behind it.

Repeating  a bit later: 824 and then 902 behind it.

 

My hardware isn't changing and each is pretty unloaded except for testing.

It seems likely that the Rogers side is variable.

 

During the last test pair, I ran mpstat on the router.  I don't know how accurate it is, but the % idle got down to almost 0 during the test from the router but stayed around 50 during the test from behind the router.  So NAT isn't eating the CPU on this machine.  But the R7000's processor may not be as efficient.

 

The cost of NAT is probably per-packet rather than per-byte.  The MTU is 1500 so the pMTU to speedtest is probably 1500 so the R7000 has close to 1500 * 8 cycles on each of two processors to handle a packet (1Gb/s cancels 1GHz).  That seems like a lot but memory access is likely a bottleneck.

 

I have no idea whether the R7000 firmware has or can easily have the mpstat command or something like it.  It's kind of nice to run a full distro on a router (but it may be a bit of a security risk due to the large attack surface).

Re: Rogers Hitron CODA-4582 Hardware

arnym21
I Plan to Stick Around

@HughR wrote:

@Datalink

So NAT isn't eating the CPU on this machine.


Not sure what this means, but your Linux PC-router was at 50% CPU load merely passing unaltered packets flow at near G wired speeds. Given no OS telemetry and extremely light well network optimized Linux OS, that's pretty high load. With Windows 10 telemetry, which aggressively reads all incoming data and writes to disk all the time if you care to monitor it, and a large number of various services and processes running, that translates to higher CPU load on a Windows PC. However, I don't think limitation of speeds on a Windows PC is only due to older CPU, but due to buses practical limitations too and all chipsets involved.

 

From cost standpoint, there is no advantage of running a standalone PC as a router unless it performs a significant network protection & packet analysis function, since consumer routers have quite effective hardware accelerated packets processing, and switching is usually done by a separate chipset. One may argue about better USB3.0 support on a PC-router, but again it depends on HDD file system, its drivers, and USB drivers used.

Re: Rogers Hitron CODA-4582 Hardware

Nope, going to disagree on this one "since consumer routers have quite effective hardware accelerated packets processing".  We keep running into this again and again.  This doesn't work as new users think it should and is simply not true for some, if not a good majority of routers.  Hardware accelerated packet processing, at least in terms of straight throughput, is disabled on Broadcom based routers when any of the following functions are selected, QOS, Traffic Monitoring, Port Forwarding (?) and a few others that don't come to mind at the present time.  Cut Through Forwarding is incompatible with those functions, and has been for a good many years.  There isn't any debate about this issue at this point in time.  When NAT Acceleration (CTF) is disabled you lose the fast path thru the router as the packets are routed to the processor, taking an initial througput hit, and then you take another throughput hit due to the user selected processing that occurs.  So, the choice that one has to make, as the internet plans step up in data rates, is to run the router for speed, or functions.  If you run a Broadcom based modem then you have to make a decision, speed or functions.  If you want to run gigabit rates and run packet scans or any other type of user selected or security function, then processor horsepower is the only way that your going to be able to do this.  You can't have both on a Broadcom based router.  A pc based router might not make sense economically, but, if you're looking for throughput capability, security, and user selected functionality, that is most likely the only way that your going to achieve it.  Just for heck of it, look at the Asus RT-AC5300, which runs a new Broadcom 1.8 Ghz 64 bit processor.  The 1.8 Ghz rate isn't tremendously fast, but, maybe the 64 bit capability will result in fast performance if you want gigabit throughput plus functionality.  For $449.99, you could probably build or buy a reasonable pc, running PfSense, OpenSense or Sophos and end up with wired performance that would beat any all-in-one router on the market, considering throughput and running security plus other functions. 



Re: Rogers Hitron CODA-4582 Hardware

HughR
I Plan to Stick Around

@arnym21 wrote:

@HughR wrote:

@Datalink

So NAT isn't eating the CPU on this machine.


Not sure what this means, but your Linux PC-router was at 50% CPU load merely passing unaltered packets flow at near G wired speeds.

I'm not sure what my observations mean either. Performance analysis is quite tricky. I have not been thorough.  mpstat might be misleading. With very light duties (AFAIK) the idle % jumps around between 57 and 80 (right now).

Ahh: top says that firefox, doing nothing but sitting on the speedtest page, is eating 63% of one CPU and 60% of the RAM.  Sheesh!  When I kill it, the mpstat says the system is over 99% idle.

I know that guessing about performance has very low reliability. It is useful for hypothesis forming but that's about it. So I thought I'd contribute a rough and ready measurement.

 

Given no OS telemetry and extremely light well network optimized Linux OS, that's pretty high load.

I don't much care for that word "telemetry". It's jargon introduced by Microsoft to sanitize surveillance. And (we hope) that doesn't happen in the R7000 either. It uses a Linux stack too. (I would guess that NAT hardware acceleration is hacked in in an ugly way and is not upstreamed.)

With Windows 10 telemetry, which aggressively reads all incoming data and writes to disk all the time if you care to monitor it, and a large number of various services and processes running, that translates to higher CPU load on a Windows PC. However, I don't think limitation of speeds on a Windows PC is only due to older CPU, but due to buses practical limitations too and all chipsets involved.

I don't run Windows seriously and don't care too much about it. My impression is that Anti-Virus software is the biggest performance sink. But the funniest things seem really slow. A little TV box has spent most of today at 29% of the way through an update.

As I understand it, Window's IP stack started out as a copy of the BSD stack. That should be roughly as good as Linux's. Of course code can get crudded up after years of maintenance. On the other hand, I would bet the stack has been subjected to intense performance improvement.

PC hardware is likely a lot faster than a cost-reduced design of a consumer router. Busses are a lot wider in a PC. Caches are a lot larger. Processors are usually faster (faster clocks, more work per cycle).

 

From cost standpoint, there is no advantage of running a standalone PC as a router unless it performs a significant network protection & packet analysis function, since consumer routers have quite effective hardware accelerated packets processing, and switching is usually done by a separate chipset. One may argue about better USB3.0 support on a PC-router, but again it depends on HDD file system, its drivers, and USB drivers used.


 

I"ve been building my routers out of PCs for about 20 years. The reasons are functionality and security. Not cost. They've run open source (Linux) long before any router firmware source code was released. Some of the code I've written has made it into some commodity routers but it ran on mine first.

Re: Rogers Hitron CODA-4582 Hardware

arnym21
I Plan to Stick Around

@Datalink

 

Your "disagree" statement seems significantly broader in scope and narrower in HW choice than the boundaries you outlined in the quotes. Smiley Happy I didn't limit the scope to always overheating Broadcom based routers in particular, while the solution works very well in my observation showing near 0% load at min processing and close to G wired speeds with custom FW with some other models. Interesting is, some custom FW better and deeper controls relevant chipset functionality than factory FW. Your excurse towards PC-router also complements what I said rather than opposes it, since I mentioned there is NO advantage to use a PC-router unless it carries in-depth packet inspections, usually covered by functionality  of these packages you mentioned.

Re: Rogers Hitron CODA-4582 Hardware

arnym21
I Plan to Stick Around

@HughR wrote:
My impression is that Anti-Virus software is the biggest performance sink.


Its quite difficult to find out where the network performance sink is on a Windows PC, I'm still trying. As you said it looks like everything is broader spec wise, but... it just won't add up to G network speeds in Windows, if your HW is not the last gen. Its possible that numerous chipsets drivers aren't optimized to work together, since on Mac network seems to work faster even with older HW, signaling better integration. Your TV seems to suffer from severe memory and CPU power deficiency, plus apps aren't updated for some older brands. Smiley Embarassed

Re: Rogers Hitron CODA-4582 Hardware

HughR
I Plan to Stick Around

@arnym21 wrote:
I didn't limit the scope to always overheating Broadcom based routers in particular, while the solution works very well in my observation showing near 0% load at min processing and close to G wired speeds with custom FW with some other models. Interesting is, some custom FW better and deeper controls relevant chipset functionality than factory FW.

Oh, you have measurements.  That's great!  Are they written up somewhere?

 

As I understand it, the hardware for things like NAT acceleration tend to be designed in a way that makes them very difficult to use in a way that respects the layering of network stacks.

 

In the world of big routers, functionality is broken down into the data plane and the control plane.  The control plane is implemented with CPUs.  The data plane is implemented with FPGAs, ASICs and the like and runs at circuit speed.  The IETF tries to design protocols to fit into this model.

 

My home network evolved from 10base2 (AKA thinnet) to 100baseT to 1000baseT.  The next step may require more hardware assistance within the router.  Some NICs allow offloading work from the stack (I don't know anything about their architecture).

Re: Rogers Hitron CODA-4582 Hardware

Nick_W
I Plan to Stick Around

I'm hoping someone here can help me out.

 

I just switched from the 250u package to the gigabit package, and switched my old modem/route to the Hitron CODA-4582.

 

My old modem/router was in bridge mode, so my plan was to do the same with the new CODA. Unfortunately my local network is all 192.168.100.x addresses, and in bridge mode, the CODA local port sets itself up with the address 192.168.100.1. This conflicts with my local net, and so will not work. This 192.168.100.1 address seems to be hard coded (I can't find where to change it anyway), and I'm not about to change my whole network scheme to get round this problem.

 

My speed has gone from 250MBps (ish) to 370MBps (ish), but this is not gigabit! I do get 30MBps upload though.

 

I'm having to use the CODA as a router with WiFi disabled (I have a Netgear R9000 X10 router as my primary, and a R7000 in AP mode attached to it). My server is running two 1GBps interfaces bonded together to the R9000 (which also supports bonding), so my local speeds (hardwired) should be good.

 

I have disabled IP6, put my R9000 in the DMZ on the CODA (so the internet can see it), but no luck on improving speeds. I'd like to try bridged mode - does anyone know if there is a way to change the seemingly hard coded 192.168.100.1 address the CODA has in bridge mode? or another way to improve my download speeds?

 

Thanks.

Re: Rogers Hitron CODA-4582 Hardware

Nick_W
I Plan to Stick Around

Just to update this, finally hooked my laptop up hardwired, and I’m getting speeds of about 930Mbps on all the web based testers (on any port in my network), so the speed is actually good.

 

Ive concluded that the command line speed test results are not accurate for high speeds (my servers are headless).

 

does anyone know of an accurate cli Speedtest? Speedtest.py and Speedtest-cli dont give accurate results.

 

my problem with the hard coded ipnaddress still stands though. Anyone know how to change the ip of the CODA in bridge mode?

Re: Rogers Hitron CODA-4582 Hardware

Webslinger
I Plan to Stick Around

Hello,

 

Why am I being told by a Rogers Tier 2 rep that my signal/noise ratio on channel 32 is perfectly fine? He claims anything over 25 is fine. Is that true? His solution is to replace my modem. How is replacing my modem going to fix noise coming into my line? Sometimes channel 32 (frequency 843000000) doesn't even appear in the list.

 

CODA-4582U

Hardware Version1A
Software Version2.0.10.28T2

 

Downstream Overview
Port IDFrequency (MHz)ModulationSignal strength (dBmV)Channel IDSignal noise ratio (dB)
1591000000256QAM-1.000738.605
2849000000256QAM-2.500236.610
3855000000256QAM-3.000337.356
4861000000256QAM-2.400437.356
5579000000256QAM-0.600538.605
6585000000256QAM-1.100638.605
7303000000256QAM0.000138.983
8597000000256QAM-1.200838.605
9603000000256QAM-1.100938.605
10609000000256QAM-1.0001038.605
11615000000256QAM-1.1001137.636
12621000000256QAM-0.8001237.636
13633000000256QAM-2.9001337.356
14639000000256QAM-4.1001436.387
15645000000256QAM-5.3001536.387
16651000000256QAM-6.2001635.780
17657000000256QAM-5.5001735.780
18663000000256QAM-3.6001837.356
19669000000256QAM-2.6001937.356
20675000000256QAM-1.6002037.636
21681000000256QAM-0.5002138.605
22687000000256QAM-1.0002238.983
23693000000256QAM-0.8002338.605
24699000000256QAM-0.8002438.605
25705000000256QAM-1.2002538.983
26711000000256QAM-1.7002638.983
27717000000256QAM-2.9002737.636
28723000000256QAM-2.9002837.356
29825000000256QAM-1.3002937.356
30831000000256QAM-1.6003037.936
31837000000256QAM-1.9003137.636
32843000000256QAM-0.9003227.607
OFDM Downstream Overview
ReceiverFFT typeSubcarr 0 Frequency(MHz)PLC lockedNCP lockedMDC1 lockedPLC power(dBmv)
0NANANONONONA
14K275600000YESYESYES-0.700001
Upstream Overview
Port IDFrequency (MHz)ModulationSignal strength (dBmV)Channel IDBandwidth
130596000ATDMA - 64QAM38.00046400000
238596000ATDMA - 64QAM41.25063200000
323700000ATDMA - 64QAM37.00056400000
OFDM/OFDMA Overview
Channel IndexStatelin Digital AttDigital AttBW (sc's*fft)Report PowerReport Power1_6FFT Size
0DISABLED0.50000.00000.0000-inf-1.00004K
1DISABLED0.50000.00000.0000-inf-1.00004K

 

 

Re: Rogers Hitron CODA-4582 Hardware

ggparts
I Plan to Stick Around

Still getting disconnected using our new Rogers Hitron CODA-4582

 

or should I say the gigabit gateway refuses to assign an IP address to my wireless devices frequently.

Randomly and without warning we will go from connected to the internet to not.

I have read almost 163 pages of the Rogers Hitron CODA-4582 thread.

Wired devices seem OK.

My Lenovo laptop can be used as a "hotspot" and the iPhone will connect to that every time.

I have assigned an IP address to my S8 and that seems to have MOSTLY resolved the problem.

We were having some Signal strength (dBmV) issues (too high) the tech cam an added a splitter to reduce the signal going to the modem (-7db) and he was convinced all would be fine, but alas it is not.

the tired and true resolution (temporary only) is the unplug and replug the modem.

What does this actually do? 

It seems that all is well until we leave home (disconnect from the network) and upon return often times I will need to "reset" the modem because one or there of us cannot use the internet.

When the tech was there I showed him the list of errors (more than 20) under events I think.

He asked that I reset it and I did, (Saturday Nov 18 5pm) the list is now populated with 20 more some titled critical.

I am hoping that the fact that it seems better when we unplug then replug means something specific that I can look at?  I want to say the FW is quite new as we have had the modem like 7 days. 2.0.10.28T2

 

Re: Rogers Hitron CODA-4582 Hardware

ggparts
I Plan to Stick Around

Here are my event logs from today is anything specific indicated?

 

The DOCSIS event logs is shown here

No. Time Type Priority Event
1 11/21/2017 08:05:28 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
2 11/21/2017 08:06:04 84020200 warning Lost MDD Timeout;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
3 11/21/2017 08:26:35 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
4 11/21/2017 08:26:35 82000300 critical Ranging Request Retries exhausted;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
5 11/21/2017 08:26:35 82000600 critical Unicast Maintenance Ranging attempted - No response - Retries exhausted;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
6 11/21/2017 08:26:36 82000300 critical Ranging Request Retries exhausted;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
7 11/21/2017 08:26:36 82000600 critical Unicast Maintenance Ranging attempted - No response - Retries exhausted;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
8 11/21/2017 08:58:06 82000400 critical Received Response to Broadcast Maintenance Request, But no Unicast Maintenance opportunities received - T4 time out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
9 11/21/2017 08:58:24 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
10 11/21/2017 08:58:36 82000400 critical Received Response to Broadcast Maintenance Request, But no Unicast Maintenance opportunities received - T4 time out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
11 11/21/2017 08:58:56 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
12 11/21/2017 09:26:36 82000400 critical Received Response to Broadcast Maintenance Request, But no Unicast Maintenance opportunities received - T4 time out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
13 11/21/2017 09:26:57 85010200 warning TCS Partial Service;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
14 11/21/2017 09:27:05 82000400 critical Received Response to Broadcast Maintenance Request, But no Unicast Maintenance opportunities received - T4 time out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
15 11/21/2017 14:28:18 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
16 11/22/2017 01:58:51 90000000 warning MIMO Event MIMO: Stored MIMO=-1 post cfg file MIMO=-1;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
17 11/22/2017 18:34:31 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=aXXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
18 11/23/2017 09:11:46 90000000 warning MIMO Event MIMO: Stored MIMO=-1 post cfg file MIMO=-1;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
19 11/23/2017 12:05:06 82000200 critical No Ranging Response received - T3 time-out;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;
20 11/23/2017 16:22:29 90000000 warning MIMO Event MIMO: Stored MIMO=-1 post cfg file MIMO=-1;CM-MAC=XXXXXXXXXXX;CMTS-MAC=00:17:10:90:da:12;CM-QOS=1.1;CM-VER=3.1;

Re: Rogers Hitron CODA-4582 Hardware

ggparts
I Plan to Stick Around

Anything here?

I am grasping as the family is on me every day since the upgrade

Downstream Overview
Port IDFrequency (MHz)ModulationSignal strength (dBmV)Channel IDSignal noise ratio (dB)
1597000000256QAM7.700836.387
2849000000256QAM7.500235.595
3855000000256QAM7.000335.418
4861000000256QAM6.600435.084
5579000000256QAM7.400536.387
6585000000256QAM7.600636.610
7591000000256QAM7.600736.387
8303000000256QAM1.000136.387
9603000000256QAM7.400936.610
10609000000256QAM7.7001036.387
11615000000256QAM7.4001136.610
12621000000256QAM7.2001236.387
13633000000256QAM7.1001336.387
14639000000256QAM6.3001435.780
15645000000256QAM6.6001536.387
16651000000256QAM7.3001636.387
17657000000256QAM7.7001736.610
18663000000256QAM7.7001836.387
19669000000256QAM7.2001935.780
20675000000256QAM7.3002036.387
21681000000256QAM7.1002135.780
22687000000256QAM7.7002236.610
23693000000256QAM7.6002336.387
24699000000256QAM7.9002436.610
25705000000256QAM8.0002535.780
26711000000256QAM7.7002635.084
27717000000256QAM7.8002733.377
28723000000256QAM7.7002832.237
29825000000256QAM7.2002935.595
30831000000256QAM7.2003035.595
31837000000256QAM7.5003135.780
32843000000256QAM7.1003235.595
OFDM Downstream Overview
ReceiverFFT typeSubcarr 0 Frequency(MHz)PLC lockedNCP lockedMDC1 lockedPLC power(dBmv)
0NANANONONONA
14K290600000YESYESYES2.500000
Upstream Overview
Port IDFrequency (MHz)ModulationSignal strength (dBmV)Channel IDBandwidth
130596000ATDMA - 64QAM40.00016400000
238596000ATDMA - 64QAM43.25033200000
323700000ATDMA - 64QAM40.00026400000
OFDM/OFDMA Overview
Channel IndexStatelin Digital AttDigital AttBW (sc's*fft)Report PowerReport Power1_6FFT Size
0DISABLED0.50000.00000.0000-inf-1.00004K
1DISABLED0.50000.00000.0000-inf-1.00004K

Re: Rogers Hitron CODA-4582 Hardware

FITIMS
I Plan to Stick Around

Hello Rogers Community.

 

Here is my issue, hopefully someone can help me out... I will try and keep it short.

I used to be in the 250u Plan before (250down/20up) for the past two years.

 

It was time to get a better deal which I thought I am getting by upgrading to 500U plan.. 500mb down speed.

With it of course came along the Coda4582 as you all may be aware.

 

All fun and exciting for us who love internet and are a bit techy...until i plugged that modem in and noticed the speed was nowhere near 500. With the older (Advance Wifi Modem) I was getting consistent speed on wifi from any device (laptop/imac/tablet/tv/etcc)... with this Coda modem... horrible.. one device has 50mb, one has 200, one has 100 one has 300??..

 

I have called and got a tech in and its been an on-going issue and they pretty much ended up with .... WELL you are paying for a wired speed not wireless speed.... like Who The ---- has a wired computer nowadays?... everything is wireless.. I have been a rogers customer for over 15 years and I am ready to leave any minute if this doesnt get resolved asap. Been patient for a month now...

 

I think the new modem CODA-4582 is the issue... Wired, i do get good speed Wifi is bad compared to my older modem.

 

Can anyone help on this? Please and Thank you

Re: Rogers Hitron CODA-4582 Hardware

@FITIMS can you have a look at the following post regarding modem wifi settings and looking at one's wifi environment and see if setting and channel changes might resolve some of the slow wifi issues.  Note that beamforming is not enabled on the modem.  That will probably increase the receive signal levels by a dbmw or two, but, that alone might help increase the data rates by one or two levels on the modulation and coding scheme:

 

http://communityforums.rogers.com/t5/Internet/slow-wifi/m-p/398557#M46745

 

http://mcsindex.com/



Re: Rogers Hitron CODA-4582 Hardware

FITIMS
I Plan to Stick Around

@Datalink Thanks for your reply.

I have followed that guide from that link you sent me and made the changes as suggested for both 2G and 5G and the results are the same.

Although I couldnt find the "beamforming" option when I logged inside the modem to enable it? or maybe I missed it, but I did check couple times. (maybe i missed it?)

 

The app on the link, WIFI Analyzer, doesnt really show me anything, unless I dont know what to look for, but I am looking if I can see what channels are occupied by neighbours which I cannot seem to find.

 

Please advise,

 

Thanks in advance.

Re: Rogers Hitron CODA-4582 Hardware

@FITIMS, beamforming is not enabled in the 4582 modem and there are no plans to enable it.  So, as a result, there is no user option available either.  That is normally found on routers which use 802.11ac wifi, so, if you were to purchase a router, that would or should be available for use.

 

For the channel selection, I suspect that you are able to see numerous 2.4 Ghz networks that are present in your neighborhood.  The problem in that case is to determine which of the channels 1, 6, or 11 have the least amount of competition, both in terms of the receive power level from other networks which might compete with your networks, and then number of networks on any of those channels.  Ideally, for both the 2.4 and 5 Ghz networks, you would have somewhere around 40 to 45 dBmW, or greater separation in power levels between your network receive power level and that of the nearest competitors.  At that amount of power separation, you can expect your network to operate normally.  As that power level separation is reduced, depending on how far away those other networks are from your home, you will see increasing difficulties in running your own network.  So, looking at a wifi analyzer, you're looking for a 2.4 channel, normally 1, 6, or 11 that offers that combination of fewer competitors and greatest level of separation in power levels from your network to the nearest other network.  The same applies for the 5 Ghz channels in the channel 149 to 161 range that offers the same.  Since the 5 Ghz channels have a shorter range you should see far less networks in the display.  

 

In terms of running a wifi analysis my recommendation would the following devices to run the analysis:

 

1.  The modem's built in wifi scanner

2.  A laptop running inSSIDer

3.  A smartphone app

 

The reasoning in this case is that the modem has three 2.4 antenna and four 5 Ghz antenna.  A laptop usually has two antenna, sometimes three in the case of a higher end MAC, and sometime only 1, in the case of an everyday laptop.  The smartphone might have one, maybe two antenna.  What this all boils down to is the fact the number of antenna will determine how sensitive the device is, the greater the number of antenna, the better receive and transmit capability it will have, including the ability to see other networks that perhaps a single antenna smartphone or laptop can see.  

 

If as you indicate, you can't see other 5 Ghz wifi networks, that could mean that you're in luck and there is no competition in the channel 149 to 161 range.  Or, depending on the device that you are using, it might not be very sensitive to other network if its only running a single antenna, or, perhaps the device doesn't support 5 Ghz networks.  I'm thinking specifically of laptops that only have 2.4 Ghz capability, and there are a good number of them around.  We keep running into problems on the forum when new customers arrive with equipment that only supports 2.4 Ghz networks, much to their surprise and disappointment.  If you fire up the Wireless Survey, which is located in the ADMIN .... DIAGNOSTICS tab, you should be able to very quickly determine if there are other 5 Ghz networks running nearby.  The data is text only, but, if you copy that data, you can dump it into something like MS Excel and expand the channel and received power level data to determine if there is any other nearby competition and what channels in that 149 to 161 range are open.  Typically there is a base channel, and extended channels.  Users set the base channel and the modem or router takes care of the extended channel management.  Any scanner should show the base channels that are in use. 

 

Hope this helps. 



Re: Rogers Hitron CODA-4582 Hardware

FITIMS
I Plan to Stick Around

@Datalink Thanks for the explanation.

I got an app for my Imac and Believe me I do have a lot of competition. More in the 2G than 5G. But even in the 5G there is still a lot of them. I also did the Wifi Survey and here it is...

 

Re: Rogers Hitron CODA-4582 Hardware

What you will end up doing is competing with Kitty for Channel 157 air time.  When a modem/router/user device is done with a channel, there is a pseudo random timer that runs in every device.  The first device to time out wins the channel.  That would occur in the case where the wifi receiver recognizes that the channel is occupied, even if it can't decrypt the data.  In the case of the other users for channel 149, there is a chance that the wifi card will simply ignore them due to the power levels.  The power level in this case puts 0 dbmw at the top and descends from there, so an increasing negative power level indicates less received power from that transmitter.  Typically the best you will do from your own modem to device is in the range of -40 to -45 dbmw.  I would say that Kitty must be your next door neighbor, from looking at the received power level on the modem.  

 

So, the question comes down to the results on Channel 157, with nearby competition, versus running on 149 or 161.  I would set the bandwidth to 80 Mhz and let the modem, in conjunction with the user device, determine what extension channel is in use for each broadcast.  Run a speedtest, with 80 Mhz bandwidth for channels 149, 157 and 161 and choose the best of the three.  Run that test at the location where you would typically use your laptop/tablet/smartphone.  Use the www.speedtest.net Toronto or Montreal Rogers servers, depending on which is closer to your home.  Those servers have the horsepower required to run gigabit speedtests .  In terms of using 80 Mhz wide channels, for each broadcast, there is a clear channel check that is carried out by the receiving device.  That check is requested by the transmitting device but, its actually done at the receiving end.  Simply, the receiver determines if any of the extension channels required for 40 or 80 or 160 Mhz is occupied.  If any of the extension channels are occupied, the transmitter will default down to 20, 40 or 80 Mhz wide channels.  It depends on the user selected channel bandwidth and the results of the clear channel check.  This is done for every transmission, so, it changes on the fly.  

 

Fwiw, I'm also competing with my neighbors in that same channel 149 to 161 range.  Despite that competition, I get better data rates, with the higher power outputs in that channel range than I do with the lower channels starting down at channel 36, which only run 50 or 200 mw for output power versus the 1 watt at the higher channel ranges.  In the lower channel range, for my location, I have far less competition.  Despite that, our wifi rates are higher in the higher channel ranges.  To know that, one would have to run a test for both the low and higher channel ranges and run that test at a location where the devices in question are typically used.  There will be a varience in the results as a result of the physical range from the modem to the user device.   At some point, if you typically run the devices at closer ranges, you might find that it makes no difference between the low and high channel ranges, in terms of the speedtest results.

 

Some additional food for thought, I'm not sure if its available in your firmware version.  In the WIRELESS .... ADVANCED settings, you might find a DFS setting.  If its there, and you enable that, you will have access to the mid channel range between the low and high 5 Ghz channels.  You might be the only one of the block.  Here's a link to a site that shows the Dynamic Frequency Selection channels in Canada:

 

https://www.semfionetworks.com/blog/5ghz-band-channel-availability-in-canada

 

 

If that Dynamic Frequency Selection is available and you want to try it out, have a look at the following post:

 

http://communityforums.rogers.com/t5/Internet/CODA-4582-Open-Issues-for-Investigation/m-p/397331#M46...

 

I would try both 80 and 160 Mhz wide channels.

 

Unfortunately running a wifi network isn't just a matter of setting it up and letting it run itself. There is some level of knowledge required to fine tune the wifi settings and determine where the best operating channel might be, and that might take some experimentation. Changing the wifi settings will get you out of the slower rates that can occur with the default settings. From there, its a matter of determining what channel to use.

 

At the end of the day, if you live on wifi, strongly consider buying a router and running the modem in Bridge mode.  My personal prefernce is for routers with external antenna.  I would advise anyone running 500 Mb/s and higher to look for a router with the fastest processor available.  To run higher data rates and any security features, and or plus VPNs, and other features will take a considerable amount of horsepower, compared to what we traditionally think of in terms of routers.  So, strongly consider whats available with 1.8 Ghz procesors and go no lower than 1.4 Ghz processors.  Also look for external antenna, which gets the antenna off of the motherboard and away from any shadowing effects that can be caused by the motherboard components.