@WestPoint wrote:@RogersDave
According to the change log in this acutall thread, and also to the gentleman on rogers Teir 2 Tech Support, who helped to REMOVE 28T2 from a friends CODA last week and join the the trial.....
28T2 doesnt exist. never mind released or the standard.
How then the CSR was able to remove 28T2 if it doesn't exist? 
If you had CODA you'd find by login to its GUI it comes with 28T2 preinstalled from factory, latest probably shipped with 32 as Dave said. Trial FW was released over time under various builds, some of them became production later shipped by Hitron, but original FW they developed themselves. If you reset the modem it'll revert back to its factory FW.
@arnym21 there is no speed restriction for the 4582, in terms of the modem operating in Gateway or Bridge mode. You should see the same speedtest results from either mode. Why you don't see that is mystifying to say the least. As you indicated earlier, installing a different NIC might make a difference. Beyond that, the processor speed, memory amount on the motherboard and hard drive type and transfer rate also comes into the picture.
One thought that came to mind, you will see different WAN IP addresses when you shift the modem from Gateway to Bridge mode. The CMTS assigns IP addresses which are in part based on the device MAC address. The IP address are not static and do not appear to be assigned statically to a specific port at the CMTS which then connects to the modem. If you run the modem in Bridge mode, you can actually change the routers MAC address by cloning different MAC addresses. However, you would have to ensure that the MAC address you are using isn't already in use at your CMTS. The router will let you know if the MAC address that you are attempting to use is not a valid MAC address. I'm wondering if the reason that you see a speed drop is partly or wholly based on a MAC address issue?
Keep in mind that although one can run ~950Mb/s on a download test, that doesn't mean real world sites are going to give you the same data rates from their servers. So, in a way the speedtests are artificial and really prove that the "last mile" from the CMTS to the modem, and your equipment can support rates up to and including 950 Mb/s on a gigabit plan.
Two points come to mind:
1. Its always possible that you have a bad modem. That's actually pretty rare from what we see in the forum, but, you could swap it out for test purposes. That would eliminate the modem from the list of causes; or
2. There is a configuration issue with your CMTS. If so, this is the first issue of its type that has been posted in the forum, where one modem mode works and the other doesn't. That would require someone like @RogersDave to check your CMTS configuration.
Good suggestions. Lets hope they'll look into that before the next trial FW build.
As to "real world sites", interesting point, especially when you discuss it with Bell Fiber Gigabit subscribers who wows their real world browsing experience. 
Hello Community,
I just received some fresh news from Hitron. They believed they have identified the source of the reboot issue and will be providing me an incremental build tomorrow (I'm expecting it to be 2.0.10.33T2).
As soon as I have this firmware, I will push it to the firmware trial participants who identified this as an issue in this thread (@tester2013, @JohnBeaudin, @hoopdogg, @zardoz99, @lethalsniper, @Windwalker, @pacerdjl, @RyzenFX, @georgic, @Nadernt, @Triple_Helix, @Mythen, @User1445713, @sputnik62, @arnym21, @WestPoint). I'll then proceed with all the other modems Thursday and Friday based on stability.
Some other answers
@WestPoint, the firmware 2.0.10.28T2 does exist and is really the production one. I believe that frontline agents however don't see it labelled as 28T2 but only 28. The reason it is not in the release notes in this thread is because this firmware was never tested as part of this firmware trial program. It is essentially the same as firmware .27 without Rogers branding.
@arnym21, Factory reset will make the modem accept a full configuration from the network. This configuration states that the modem must be running network firmware 2.0.10.28T2 (which the modem will download automatically).
Regarding the IP pools between Bridge modem and Gateway modes, they are not different. What happens is that each CMTS has multiple IP pools from very different ranges (it occurred over the years as we added IP blocks). When a new MAC address is seen on the network, or a MAC address we haven't seen in the last seven days, our DHCP server pick an available IP and assign it to the device. In gateway mode, the MAC address that our DHCP server sees is the modem MAC while in bridge mode, it will be the MAC of the 3rd party router. Switching from bridge to gateway therefore results in an IP change but they could technically be in the same range, it is random.
The QoS configuration also is not tied to the IP address but to your modem class of service which is linked to the plan you subscribed to. For example, if you subscribe to Ignite 500u, the class of service will cap the speed at 500 Mbps downlink and 20 Mbps uplink. There is no difference between the class of service in bridge or gateway mode.
In your specific case, if your speed is reduced when in bridge mode compared to gateway, I suspect it has to do with the capability of your 3rd party router but you'd have to provide some specifics on the model of router you are using. We have experts in this community that should be able to assist in optimizing the settings for the best performance.
--Dave
@georgic, there might be other issues at play here. The CODA modem MoCA channel is only enabled to run in the D-High band, which runs from 1350 to 1675 Mhz. That is to allow the Rogers tv modems to run MoCA for Whole Home PVR. A bonded channel consists of two 100 Mhz wide channels with a 25 Mhz gap inbetween. So, if the bonded channels start at 1350, that results in the upper edge of the second bonded channel sitting at 1575 with a 100 Mh tuning range above that. The MoCA 2.0 spec indicates that the bonded channels run in the D-High sub-band. Just to note, what I haven't found is an indication within the MoCA 2.0 spec to indicate that the bonded channels can run anywhere in that band and are not necessarily tied to the bottom starting frequency of 1350 Mhz. As you will see below, that's possibly a moot point.
Looking at the specs for the Antronix splitters which Rogers uses, there is no mention what so ever of supporting any MoCA operation. If it actually works, room to room, that might be a minor miracle rather than good planning. The upper frequency tops out at 1002 Mhz, far below any MoCA operating range. The question in this case is how fast the frequency roll-off occurs above 1002 Mhz. Looking at the amplifiers, there is only one amplifier that is shown to support MoCA 2.0 operation with the 1675 Mhz high end, but, the low frequency split for the upstream frequencies is not correct for Rogers current frequency allocations. The other amplifiers are only MoCA 1.1 rated which tops out at 1500 Mhz.
So, this is an interesting conundrum. The modem MoCA 2.0 D-high band runs from 1350 to 1675 Mhz. The splitter that you might have installed doesn't appear to support MoCA operations. If you happen to have a MoCA Qualified amp installed, from Antronix, it doesn't support MoCA 2.0. So, if the Modem MoCA firmware is enabled to run Turbo Mode, and I don't see why it wouldn't be, with the current splitters and amps in use by Rogers, it would appear that Turbo Mode is not achievable unless you swap you're splitter or amp for one from another manufacturer such as Holland Electronics which manufactures MoCA 2.0 splitters and amps.
So, here are some points to ponder:
- What is the real frequency response of the splitters and amplifiers installed by technicians? Is that response really as indicated on the web site? Is there an updated spec which isn't shown on the Antronix website? I've been keeping my eyes open for any techs visiting the neighborhood to inquire about any datasheets they might have on hand with the splitters and amps. That would answer this question in a heartbeat. Until its been confirmed that the splitters and amplifiers actually support MoCA 2.0 frequency ranges, I wouldn't count on attaining the turbo or bonded channel performance that is shown on the modems datasheet which is:
- Demodulation/ Modulation: BPSK, QPSK, 8QAM, 16QAM, 32QAM, 64QAM, 128QAM, 256QAM, 512QAM, 1024QAM
- PHY data rate: 700Mbps (baseline Mode) / 1400Mbps (bonding channel)
- Throughput: 400+Mbps (baseline Mode) / 500+Mbps (turbo mode, point to point) / 800Mbps (bonding channel)
- Frequency range: 1125-1675MHz
- Channel bandwidth: 100MHz (baseline mode) / 225MHz(bonding channel)
- If you're not running Whole Home PVR, you might be better off running a MoCA adapter at the modem instead of using the modem's MoCA capability as the MoCA adapter wouldn't be restricted to the D-High band as the modem currently is. The adapter should have access to the entire D-Low and D-High bands which is a 550 Mhz band:
- Sub-band D-Low (DL): 1125 to 1225 MHz edge to edge (100 MHz wide)
- Sub-band D-High (DH): 1350 to 1675 MHz edge to edge (325 MHz wide)
- Guard-band between sub-bands: 1225 to 1350 MHz (125 MHz wide)
- If you are running Whole Home PVR, then from what I can see so far, the only way to run turbo or bonded modes is to ensure that your splitter or amp supports the MoCA 2.0 D-high upper frequency limit at 1675 Mhz instead of 1500 Mhz.
References:
MoCA 1.0 spec:
http://www.mocalliance.org/MoCA1/specification/MoCA_Specification_for_Device_RF_Characteristics.pdf
MoCA 2.0 spec:
Have a look at page 9 for the frequency chart.
http://www.antronix.com/antronix-Products-MoCA.php
Here's a test that might confirm the splitter or amp frequency issue if in fact your current configuration has the modem on one splitter port and the MoCA adapter on another splitter port running to another room.
Disconnect the adapter from its current location and relocate it to sit somewhere between the splitter and modem, on the same splitter port or cable that runs to the modem. In that configuration all you have is a piece of cable between the modem and the adapter. I'm assuming that the cable is an RG-6 cable. Run a speedtest in that configuration if you can. That could show a higher result if the splitter or amplifier frequency range is preventing the modem and adapter from running in Turbo or Bonded channel mode. That frequency range wouldn't have any effect on the adapter in this configuration.
Edit: change paragraph two to reflect a newer Antronix MoCA 2.0 rated amplifier.
-
931913 views
-
669 Likes
-
-
