Rogers Hitron CODA-4582 Hardware

[Double_K]

Announced 13-December-2016 by @RogersDave  http://communityforums.rogers.com/t5/forums/forumtopicpage/board-id/Getting_connected/message-id/379...

Credit to @Hybrid_Noodle

Hitron website: http://www.hitron-americas.com/product/coda-4582/

Datasheet: http://www.hitron-americas.com/wp-content/uploads/2016/09/CODA-4582-Datasheet.pdf

60W Power Supply built into unit

Pictures of Hardware Version 1A:

***Added Labels***

[Datalink]

This isn't bufferbloat with the CODA-4582.  I'm not sure of what to call this at this point in time, but its not bufferbloat.  Perhaps a matter of fine tuning and getting all of the gremlins out of the new modems. Disappointing to say the least.  

To look at where we are coming from with the Puma 6MG modem's, ie: the CGN3 series and CGNM-3552 and where it might end up with, with newer fimrware have a look at the following thread.  

http://communityforums.rogers.com/t5/forums/forumtopicpage/board-id/Getting_connected/message-id/369...

To read thru the DSLReports threads:

http://www.dslreports.com/forum/r31079834-ALL-SB6190-is-a-terrible-modem-Intel-Puma-6-MaxLinear-mist...

http://www.dslreports.com/forum/r31122204-SB6190-Puma6-TCP-UDP-Network-Latency-Issue-Discussion

This last thread is the latest discussion thread, as we wait for Intel and Arris to finish development of the next firmware series to route all IPV4 and IPV6 TCP/IP, ICMP and UDP packets thru the hardware processor/accelerator.

[JohnBeaudin]

The CODA 4582 seems to be in BETA.

I think the release was rushed due to the puma6 chipset horror story.

Looks like it will be a long journey ahead.

[HKPolice1]

---

@Datalink wrote:

This isn't bufferbloat with the CODA-4582.  I'm not sure of what to call this at this point in time, but its not bufferbloat.  Perhaps a matter of fine tuning and getting all of the gremlins out of the new modems. Disappointing to say the least.  

To look at where we are coming from with the Puma 6MG modem's, ie: the CGN3 series and CGNM-3552 and where it might end up with, with newer fimrware have a look at the following thread.  

http://communityforums.rogers.com/t5/forums/forumtopicpage/board-id/Getting_connected/message-id/369...

To read thru the DSLReports threads:

http://www.dslreports.com/forum/r31079834-ALL-SB6190-is-a-terrible-modem-Intel-Puma-6-MaxLinear-mist...

http://www.dslreports.com/forum/r31122204-SB6190-Puma6-TCP-UDP-Network-Latency-Issue-Discussion

This last thread is the latest discussion thread, as we wait for Intel and Arris to finish development of the next firmware series to route all IPV4 and IPV6 TCP/IP, ICMP and UDP packets thru the hardware processor/accelerator.

---

I'm aware of the Puma6 issue and how the trial firmware is trying to fix it, but the Puma7 doesn't seem to have the same issue IMO. Puma6 has lag spikes even when there are no transfers happening, it even pings 1-2ms over LAN.

The CODA-4582 has much more stable & lower latencies when pinging the default gateway with no active transfers (12-13ms avg) and spikes are below 40ms. LAN response times are also always below 1ms. My previous Puma6 modem would ping 15-18ms avg with spikes upto 100ms+ under the same no load conditions.

Puma7 only starts lagging during active transfers which seems like a classic case of buffer bloat. I'm sure the CMTS has something to do with it because I remember when my area used to be on the Cisco CMTS and my ping times would only jump to ~30ms even when the downstream was fully saturated. With the Casa CMTS, buffer bloat was much worse.

In any case, are Rogers engineers aware of this Puma7 issue and is there any a plan of attack on how to solve it? 

[JohnBeaudin]

the new CODA-4582 is their new baby.. I am pretty sure they will do all they can ( Rogers) to make it the best thing in the world.

[Datalink]

I would agree with that.  I'm sure that the engineers are more than aware of the various issues, just a question of tackling them one by one.  Whats the plan?  Good question. That's a question for @RogersDave, once again.  Not trying to pass the buck, but Dave is probably the one person who has the best understanding of what the issues are and what needs to be done.  And ....... who also needs a break from all of this.  So, we're all going to have to be a little patient here.

[RogersDave]

Hello Community,

Over the holidays, we have deployed in a targeted fashion firmware 2.0.10.19 to all modems that as per our network scans were in a degraded mode. I’ve also manually processed requests on this forum for this firmware.

Given that we haven’t observed any negative impact from this firmware, we have decided to make it the official firmware revision starting tonight. We will therefore start upgrading all remaining CODA-4582 modems to version 2.0.10.19.

New modems coming from the factory are loaded with version 2.0.10.13 but within 48-72 hours will be automatically upgraded to version 2.0.10.19 as well.

Furthermore, I’m getting ready to restart the firmware trial program, likely early next week with firmware version 2.0.10.20. I will update the release log as soon as I have confirmation of the fixes included in the firmware.

By running a single production version on the network in addition to one version under test will allow us to focus our energy on addressing remaining issue and making real improvements.

I know that in the past few days, there was a lot of questions sent to me. I may not have answered everything but I do read them all (mostly in the morning) and I’m using these comments to build a list of priority issues that we will be working on next. I will likely post a message tomorrow requesting examples of people affected by specific symptoms. In some cases, Intel and/or Hitron requested to take debug logs directly on some impacted modems in order to nail down a solution to each problem and I'm hoping that we will be able to find example in this community.

These problems may not look like they impact you directly but we know that they are affecting a large number of subscribers and that even if there is no apparent direct impact to you, working on these problems first will benefit everybody.

We have made a lot of progress in the last 2 weeks and I am convinced that we will make even more improvements in the coming weeks.

Dave

[JohnBeaudin]

@RogersDave

With everyone on .19 , does that mean ipv6 will be turned on soon?

[Alex4161]

@RogersDave @CommunityHelps

Thank you for all your efforts.  It is good to see someone spearhead things and try to get resolution.  Please sign me up for the firmware trial and I can help you out as needed.

Hopefully we can get better power output from the 5 Ghz and get the 2.4 Ghz band up and working without degrading the speeds.

For some reason, my upstream speeds took a nosedive and they are as follows:

This was surprising as Dec 30 around 6 AM  (before the widepread area outage) my speeds were as follows:

As you can see, there is quite a difference in speed and nothing was changed on my side. It looks like when the network went down it was never back to where it was before.

Thanks

[Webslinger]

---

@Alex4161 wrote:

@RogersDave @CommunityHelps

As you can see, there is quite a difference in speed and nothing was changed on my side. It looks like when the network went down it was never back to where it was before.

Thanks

---

@RogersDave

@Alex4161

Yeah, I'm having the exact same issue with upload (except my upload speeds look worse) right after a fiber cut in the Finch/McCowan area on Dec. 30-31st. Rogers engineers closed my ticket that opened by a tier 2 Rogers agent without contacting me about it, claiming everything is fine, when it's clearly not: http://www.speedtest.net/my-result/5935355424

I strongly suspect that if I were subscribed to the gigabyte plans, my speedtest results would be similar to mahomed's.

That's using a wired, ethernet connection running from the cable modem to a computer. I have 4 computers here. They're all showing the same horrible upload speeds.

I'm on 2.0.10.19 firmware.

Downstream Overview

Port ID	Frequency (MHz)	Modulation	Signal strength (dBmV)	Channel ID	Signal noise ratio (dB)
1	591000000	256QAM	-4.900	31	36.387
2	657000000	256QAM	-5.500	41	36.387
3	597000000	256QAM	-5.000	32	36.387
4	663000000	256QAM	-5.100	42	36.610
5	669000000	256QAM	-4.800	43	36.610
6	675000000	256QAM	-4.800	44	36.610
7	681000000	256QAM	-4.300	45	37.356
8	687000000	256QAM	-4.500	46	36.610
9	693000000	256QAM	-4.600	47	37.356
10	699000000	256QAM	-4.600	48	37.356
11	705000000	256QAM	-4.600	49	37.356
12	711000000	256QAM	-5.100	50	36.387
13	717000000	256QAM	-5.700	51	36.387
14	723000000	256QAM	-6.100	52	36.387
15	825000000	256QAM	-8.200	53	35.595
16	831000000	256QAM	-8.700	54	35.084
17	837000000	256QAM	-9.200	55	34.926
18	843000000	256QAM	-9.700	56	34.346
19	567000000	256QAM	-5.200	27	36.387
20	573000000	256QAM	-5.100	28	36.387
21	579000000	256QAM	-4.900	29	36.610
22	585000000	256QAM	-4.900	30	36.387
23	849000000	256QAM	-9.800	57	34.346
24	855000000	256QAM	-9.600	58	34.484
25	603000000	256QAM	-4.800	33	36.610
26	609000000	256QAM	-4.900	34	36.610
27	615000000	256QAM	-4.700	35	36.387
28	621000000	256QAM	-4.600	36	36.610
29	633000000	256QAM	-5.000	37	36.387
30	639000000	256QAM	-5.100	38	36.610
31	861000000	256QAM	-8.900	59	34.926
32	651000000	256QAM	-5.700	40	35.780

OFDM Downstream Overview

Receiver	FFT type	Subcarr 0 Frequency(MHz)	PLC locked	NCP locked	MDC1 locked	PLC power(dBmv)
0	NA	NA	NO	NO	NO	NA
1	NA	NA	NO	NO	NO	NA

Upstream Overview

Port ID	Frequency (MHz)	Modulation	Signal strength (dBmV)	Channel ID	Bandwidth
1	23699453	ATDMA - 64QAM	32.250	5	6400000
2	38596000	ATDMA - 64QAM	36.000	6	3200000
3	30596000	ATDMA - 64QAM	33.000	4	6400000

[AaronMT]

My CODA 4582 just updated to .19 and ran a speed-test on a wired connection. Still not hitting high above the 700Mbps mark on the Gigabit plan.

[RyzenFX]

From the past few CODA modems I had to swap out, I noticed that it seems to be affected by major fluctuations in ping. I have experienced these fluctuations on all modems I have swapped out, and even the current modem I have now, despite it not being affected by the interference issue. 

The major fluctuations have been especially notable in all the games I play. This causes me to lag and lose alot of packets, most notable in League of Legends. In League lose 11 packets/min.  The fluctuations in ping is also notable in VOIP programs such as Discord, where ping graph shows the huge ping fluctuations. 

Based on my experiences on all the CODA modems I have been on, the modems may be plagued by some sort of packet loss/ ping instability bug.

I have yet to seen my pings stabilize at a constant number and also see my packet loss decrease, which I only have seen on the CGN3ACSMR.

[Datalink]

@RyzenFX would you happen to know if your VOIP program, Discord, uses UDP?

[Mayoo]

Thanks for .19 FW !!! 

Did Factory Reset, Rebooted both Modem and Router and also cleared the Router NVRAM. Put the modem back in Bridge mode, Rebooted again

[Datalink]

Asus router, or other?

[Mayoo]

Asus RT-87U with Latest Firmware with QOS turned off

[RyzenFX]

@Datalink

Yes, Discord does mainly use UDP for "voice data transmission"

An excerpt from the developer website states that:

utilize a separate UDP-based connection for voice data transmission. Because UDP is utilized for both the receiving and transmitting of voice, your client must be able to receive UDP packets, even through a firewall or NAT

[tyreman]

Got .19 here now

http://www.speedtest.net/result/5935881932.png

good so far

on with RT AC 68U with https,nat,ipv6 enabled

[Alex4161]

@Webslinger

I ran a test for speeds and this is what I get now:

My signal levels are as follows:

Downstream Overview

Port ID	Frequency (MHz)	Modulation	Signal strength (dBmV)	Channel ID	Signal noise ratio (dB)
1	633000000	256QAM	2.400	37	35.780
2	363000000	256QAM	6.900	10	37.356
3	369000000	256QAM	7.200	11	37.356
4	375000000	256QAM	7.500	12	36.610
5	381000000	256QAM	7.200	13	36.610
6	387000000	256QAM	6.600	14	37.356
7	393000000	256QAM	6.700	15	36.610
8	399000000	256QAM	6.700	16	36.610
9	405000000	256QAM	5.900	17	37.356
10	411000000	256QAM	5.800	18	37.356
11	417000000	256QAM	6.700	19	37.356
12	423000000	256QAM	7.100	20	37.356
13	429000000	256QAM	7.700	21	37.356
14	435000000	256QAM	8.100	22	37.636
15	441000000	256QAM	7.800	23	37.356
16	447000000	256QAM	6.900	24	37.356
17	555000000	256QAM	4.500	25	36.387
18	561000000	256QAM	3.900	26	36.387
19	567000000	256QAM	3.100	27	36.387
20	573000000	256QAM	4.000	28	35.780
21	579000000	256QAM	4.100	29	35.595
22	585000000	256QAM	4.100	30	36.610
23	591000000	256QAM	5.300	31	36.610
24	597000000	256QAM	5.800	32	36.610
25	603000000	256QAM	5.600	33	37.636
26	609000000	256QAM	4.900	34	36.610
27	615000000	256QAM	4.000	35	36.610
28	621000000	256QAM	2.800	36	36.387
29	357000000	256QAM	7.000	9	37.356
30	639000000	256QAM	2.600	38	36.387
31	645000000	256QAM	3.600	39	36.387
32	651000000	256QAM	2.900	40	36.387

OFDM Downstream Overview

Receiver	FFT type	Subcarr 0 Frequency(MHz)	PLC locked	NCP locked	MDC1 locked	PLC power(dBmv)
0	NA	NA	NO	NO	NO	NA
1	NA	NA	NO	NO	NO	NA

Upstream Overview

Port ID	Frequency (MHz)	Modulation	Signal strength (dBmV)	Channel ID	Bandwidth
1	23700000	ATDMA - 64QAM	33.250	2	6400000
2	38595746	ATDMA - 64QAM	38.500	3	3200000
3	30596000	ATDMA - 64QAM	34.750	1	6400000

OFDM/OFDMA Overview

Channel Index	State	lin Digital Att	Digital Att	BW (sc's*fft)	Report Power	Report Power1_6	FFT Size
0	DISABLED	0.5000	0.0000	0.0000	-inf	-1.0000	4K
1	DISABLED	0.5000	0.0000	0.0000	-inf	-1.0000	4K

I will say that I am on the correct profile and if there are any speed issues it is because of network congestion.

This was my speedtest result Dec 30 around 6:16 am before the fibre cut

I don't think we will really see this speed again unless we get to Docsis 3.1 or until Rogers focuses on fixing their network to what we were getting before the cut.  

[goodboyr]

This was discussed earlier in this thread, but I am still a little fuzzy on it. I have the 100u service. Am I eligible to upgrade my modem from the CGN3ACSMR to the CODA-4582? And if the answer is yes, is it a worthwhile upgrade or should I just leave well enough alone and stay with what I have? I presently have no issues, and I am achieving the expected speeds and performance.

[JohnBeaudin]

If you have no issues, you might as well wait a bit until all the issues are getting resolved on the new coda ( if you are not interested in beta testing) then it would be the best decision for you.

[Alex4161]

@goodboyr

If it ain't broke, don't fix it. If you are having no issues and are happy with performance, I would not change anything.

The new hardware is for people to try to get faster speeds on the Gigabit plans. There may be improved wifi but even that is not working properly on the CODA.

The only thing I can suggest is to see what version of firmware you have on your existing modem and see if that is the latest rogers has. I think the most recent firmware is pushed automatically so just sit back knowing that you have a proper working service 🙂