Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[roseway]

After using the workaround described above, all ports are seen as stealthed (including 631).

[les-70]

  The fix does work and the redirected port can be on a different subnet so that it can't be inadvertently used in the future.

    By the way, I can't get the FTP server to work with a usb stick. To be clear I mean the "ftp" advanced menu option and not the "usb" ftp download option under "usb".  The ftp port 21 is open on the LAN but does not respond.   

   I also still find WLAN performance disappointing and I am trying different settings at the moment.

[roseway]

I also still find WLAN performance disappointing

I agree with that, and I went back to using a separate WAP.

[burakkucat]

Thank you for the latest updates, Eric & Les. I've made a note for the next time that I have a 'play' with my HG622.

[burakkucat]

Having decided that 'the time was ripe' for another 'play' with my HG622, I set up the NAT port mapping (for port number 631, originating WAN-side) as discussed above --

Mapping Name    Interface                                            Protocol    Remote Host    External Start Port    External end Port    Internal Port    Internal Host
Block-WAN-631   WAN1_INTERNET_R_ATM1_0_38    TCP/UDP                           631                          631                          631                192.168.0.254    

I then connected to a remote system that is part of the ELRepo Project and, from there, instigated an Nmap scan of all 65536 ports of the IP address at The Cattery --

[bcat@Quad ~]$ nmap -n -T4 -A -P0 -p0-65535 IP_address_at_The_Cattery

Starting Nmap 4.11 ( http://www.insecure.org/nmap/ ) at 2013-02-25 04:33 GMT
All 65536 scanned ports on IP_address_at_The_Cattery are filtered

Nmap finished: 1 IP address (1 host up) scanned in 6567.453 seconds
[bcat@Quad ~]$

Just purrfect. 

[Subsequently edited to 'tidy up' a minor formatting error, etc.]

[roseway]

 

[GigabitEthernet]

Is everyone still happy with the HG622? I don't really want to ask this question as it is for charity (so please don't take it the wrong way), but was it worth the money you paid?

[burakkucat]

In my view, yes, it was a fair exchange, Alec.

Being a Huawei device, I am quite familiar with what is available, both via the GUI and the CLI. The inbuilt stand, the four LAN ports and firmware that hasn't been subjected to Beattie's 'fiddling' is more than enough for me to rate it above the HG612.

This is my ranking of the three Huawei Home Gateways that I own:

• HG622
• HG612
• HG610

[roseway]

I haven't tried the HG610, but the other two are both top ranking for ADSL performance in my view. The slightly odd thing is how different they are: initially the HG622 connects at a significantly higher downstream speed than the HG612, and the upstream SNR margin (on a capped TT connection) is 3 dB higher in the case of the HG622. But the HG622 downstream SNR margin falls much further at night than the HG612 (a 4 dB fall as opposed to 2.5 dB for the HG612).

Ultimately I can get rather better performance from an HG612 with the target SNR margin tweaked down by about 1.5 dB than from the HG622 with little room for tweaking.

Having said all that, I'm happy with both purchases, and will no doubt swap between them from time to time.

[les-70]

 I agree that adsl2+ wise the HG622 is good.  The Wireless side does however annoy me as it ought to be a benefit.  However many routers and adapters seem to get lots of complaints which may be environment related so this may just be for me.

  With HG622 reported signal strength on adapters is good at all locations and depending on the type of adapters the stated speeds are 270Mb/s or 135Mb/s.  Lan tests are however simply erratic sometimes real transfers computer to computer of eg 30 Mb/s on 150Mb/s adaptor but sometimes  barely 1Mb/s ( including from the web) with no reported change in the wireless connection.  Also sometimes very stubborn at getting a connection at all.  Some adapters seem better than others: I have a mix of brands.  The HG622 wireless seems fast but erratic with my mix of adapters and positions near radiators or other metal things. Any advice would be welcome!!!

  In contrast connections with my usual DG834Gv4, with less reported signal strength, are always a steady 1OMb/s computer to computer and greater than 16Mb/s from the web.

 The HG622 wireless seems fast but erratic, I would be happy if it was as good as the previous wireless G.

[asbokid]

Thanks for all the feedback! 

Glad to see that the general consensus is positive about the HG622 

So far, £920 has been raised for charity from the sale of these modems. 100% of the final sale price going to the Dogs Trust.

Thank you to everyone for their support 



cheers, a

[les-70]

 By way of on going feedback. 

The HG622 seems to give roughly one thousand times, yes more than 10**3 times, the rate of bit swap as a DG834G at the same SNR. 20/min at SNR 6.0 to 60/min at SNR 1.0 !!  The HG622 CRC errors (on fast path) are lower though, about 1/min at SNR 6.0 and about 3-4/min at SNR 1.0, about half as much as the DG834G so maybe the energetic bit swapping helps.   MY SNR daily variation is about the same on the two routers.  The HG622 seems so stable at 1.0 that I will try zero SNR next! I want to see at what stage drop out occurs.

  I continue to find the Wireless and Ethernet throughput disappointing.  An attached WGR614 with its wireless G equals the HG622.

[les-70]

   On my connection the drop out point of the HG622 seems to be at or very close to zero db SNRM. It survived heavy use for 12 hours at 0.3-0.5db SNRM but resynced when the SNRM fell to 0.1db and probably below.  Based on this I will aim to ensure it stays at or above 2db at the worst times of day.  A 2db minimum SRNM seems to give a solid low error connection with a fair margin for safety .  It corresponds to a max tweak of -2db if tweaked when my connection is at its best.   Judged relative to the same error rate the HG622 is 1Mb/s faster than the DG834Gv4. 

   Given that I am not really using the router and wireless side of the HG622 at the moment, I suspect I would be as well off with an HG612.  I can't however use an HG612  at the moment due to its lack of annex m support.

[asbokid]

Some of the VDSL2 chipsets available today are being tested using a Huawei SmartAX MA5616 MSAN.

Of the six devices tested so far, those with the BCM6368 have all come out top.

Under test are:

the two revisions of ECI VDSL2 modem- Rev./I and Rev./r (both with the Lantiq VRX268 chipset).
the HG612,  HG610 and the HG622 (all with the Broadcom 6368) and
the SagemCom F@st 2764 (with an Ikanos Fusiv Vx180 chipset, same as Fritz!Box 7390)
four other VDSL2 modems are also being studied (but not yet tested) 2Wire 3600, 3800 (Ikanos) and 3801 (Coppergate) and Planet VC-230N (Metanoia chipset)

Under zero length loop conditions, all of the devices provide close to the same actual speed (110Mbps/64Mbps).   However, it is sensitivity measured by SNR that appears to distinguish the devices from each other.

The ECI Rev/I (Lantiq VRX268) is notably inferior to the newer ECI Rev/r (also Lantiq VRX268) .
The ECI Rev/r is notably inferior under the same conditions, and in terms of SNR to the Broadcom devices.
The Ikanos is notably superior to both the ECIs, but not quite as good, in terms of SNR, as the Broadcom.

In summary, the Broadcom 6368 devices are more sensitive, providing a higher SNR across the VDSL subcarrier signals  under 0-loop conditions.

Whether that superior performance will remain when attenuation and background noise are introduced artificially remains to be seen. (next stage!)

cheers, a



* All the results, so far, are here:  http://insidehuaweima5616msan.wordpress.com/2013/02/26/contrasting-zero-loop-stats-...

[burakkucat]

I have had experience of a HG612 holding sync right down to -1.9 dB SNRM, finally 'letting go' at -2.0 dB.

As of yet, I have not observed a SNRM lower than +2.0 dB with the HG622.

For the same target SNRM, the HG622 will synchronise 250 - 300 kbps higher than the HG612.