Print

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

[NewtronStar]

They could sell all the old cabs to Nigeria or somewhere, a bit like mobile phones.

That would be an interesting site to see the old 1998 BT van parked outside a PCB cabinet and the engineer working away at it in the middle of Abuja at least it's warmer yet i can see alot of service time being made for wet joints.

[KIAB]

A thought just occured - I wonder if the V41's can also fully support G.INP, its not something Ive checked.

 VDSL2 vectoring and pair bonding

http://www2.ecitele.com/OurOffering/Products/Pages/V41.aspx

[burakkucat]

VDSL2 vectoring and pair bonding

http://www2.ecitele.com/OurOffering/Products/Pages/V41.aspx

Looking at the Technical Specifications section of that page I see --

• Supports up to 256 vectored VDSL2 lines
• Allows pair-bonding services
• One vectoring engine card per system for simplified operation
• Supports a variety of network interfaces via fiber and copper
• Ruggedized and hardened, suitable for harsh weather conditions

But no mention of G.Inp. 

[KIAB]

VDSL2 vectoring and pair bonding

http://www2.ecitele.com/OurOffering/Products/Pages/V41.aspx

Looking at the Technical Specifications section of that page I see --

• Supports up to 256 vectored VDSL2 lines
• Allows pair-bonding services
• One vectoring engine card per system for simplified operation
• Supports a variety of network interfaces via fiber and copper
• Ruggedized and hardened, suitable for harsh weather conditions

But no mention of G.Inp. 

Well, that's scuppered Kitz's idea.

[Black Sheep]

Not at all ...... 

V41 DSLAM (Vectored DSLAM chassis) with VTU-C64P V3 card, will support retransmission (G.INP). 

[Ronski]

Let's hope you know that because that's what's happening 

[Black Sheep]

Would this face lie ?? 

[kitz]

I was wondering last night if something like g.inp would be mentioned in the specs as its more of a software construct than hardware.

One thing that puzzles me and has done for months with the ECI DSLAMs is that we know they are already capable of performing g.inp for downstream, so why arent they also doing it for the upstream?   
With retransmission all the hard work & number crunching and buffer storage is at the transmitting end and this is where the modem (be that the modem in the dslam or router) needs additional processing power and buffer storage.  So if the ECI dslams do have the ability and resources to be able to cope with downstream and ability to store & send retx packets to the remote modems, then it should be a doddle for it to cope with upstream because it only has to send a re-request if a packet from the EU's modem is corrupt.  [dumbed down version]


Is it possible that Openreach simply haven't enabled it because the ECI modems in their current state cant cope with the additional resources that would be required for upstream buffer storage.   I've pondered over this for a while as to why BT have never upgraded to full retransmission when we know that the chipset is capable providing the modem has sufficient hardware resources.   We know that TP-link have been able to do so with the TD-W9980s.   I suspect it possible to do so with the HH5A's,  weve seen some hacked ECI modems with open f/w capable of doing full retransmission, so what is the missing link?  Could it be down to the model type of the ECI modems?    Its clear that Openreach are/were under the impression that anyone connected to an ECI cab is going to be using an ECI modem.   

Am I missing something here?   

[kitz]

Not at all ...... 

V41 DSLAM (Vectored DSLAM chassis) with VTU-C64P V3 card, will support retransmission (G.INP). 

Thank you BS for the confirmation.     
I had thought that any DSLAM capable of vectoring should be able to perform G.INP, due to the fact that retransmission is the recommended method of error protection for vectored lines.

VTU-C64P V3 card

Interesting that you would appear to have documentation specifically discussing the linecard model.   
We know for a fact that Openreach are already using VTU-C64's.    V3 could possibly relate to the Vinax V3 chipset that is on those linecards  - we know that the existing linecards are capable of vectoring at line card level. 

The missing link for shelf based vectoring is and always has been space on the M41's backplane for a vectoring engine card.

So back to this

IMHO the easiest way to proceed would be to swap out the M41's with a V41.

..  and back to the previous discussion of whether Openreach would consider it a worthwhile investment in upgrading the M41s with V41s.     We know they appear to be dabbling installing vectoring engine cards in the Huawei's.   The cost of a V41 is more expensive than a Huawei vectoring engine   (kind of like how a PC motherboard is more expensive than any PCI type of card) and its if Openreach would consider and go to the expense of  replacing 1/3rd of their cabs with V41s.

Im trying to think of it in simplistic sense to explain the issue with the ECI's and the comparison with PC hardware is useful to try and explain the current problem with the ECI DSLAMS.    They have the technology, they have the fibre installed, they have the PC case (cab) DSLAMs are like PCs in that they have equipment that slots into them (such as line cards, CPU, network cards, etc).  The backplane is the equivalent of a PC motherboard. ...  The M41 just doesnt have space or a spare slot to take the vectoring engine card.

So if Openreach are considering sending men out to install a vectoring engine card in the Huaweis...   then the additional cost of V41's is a drop in the ocean in the grand scheme of things and the most economical solution.  Much of the other equipment in the cab such as the line cards wouldnt need replacing.    It would however mean down-time for those DSLAMS.

[Mark07]

Doesn't sound like we're going to see vectoring or even G.INP on ECI before FttN starts rolling out 

[guest]

Im trying to think of it in simplistic sense to explain the issue with the ECI's and the comparison with PC hardware is useful to try and explain the current problem with the ECI DSLAMS.    They have the technology, they have the fibre installed, they have the PC case (cab) DSLAMs are like PCs in that they have equipment that slots into them (such as line cards, CPU, network cards, etc).  The backplane is the equivalent of a PC motherboard. ...  The M41 just doesnt have space or a spare slot to take the vectoring engine card.

The equivalent analogy would be that BT bought a motherboard which only supported previous generations of cpus but not current versions as they require a physically different socket.

[Chrysalis]

I was wondering last night if something like g.inp would be mentioned in the specs as its more of a software construct than hardware.

One thing that puzzles me and has done for months with the ECI DSLAMs is that we know they are already capable of performing g.inp for downstream, so why arent they also doing it for the upstream?   
With retransmission all the hard work & number crunching and buffer storage is at the transmitting end and this is where the modem (be that the modem in the dslam or router) needs additional processing power and buffer storage.  So if the ECI dslams do have the ability and resources to be able to cope with downstream and ability to store & send retx packets to the remote modems, then it should be a doddle for it to cope with upstream because it only has to send a re-request if a packet from the EU's modem is corrupt.  [dumbed down version]


Is it possible that Openreach simply haven't enabled it because the ECI modems in their current state cant cope with the additional resources that would be required for upstream buffer storage.   I've pondered over this for a while as to why BT have never upgraded to full retransmission when we know that the chipset is capable providing the modem has sufficient hardware resources.   We know that TP-link have been able to do so with the TD-W9980s.   I suspect it possible to do so with the HH5A's,  weve seen some hacked ECI modems with open f/w capable of doing full retransmission, so what is the missing link?  Could it be down to the model type of the ECI modems?    Its clear that Openreach are/were under the impression that anyone connected to an ECI cab is going to be using an ECI modem.   

Am I missing something here?   

It could be as simple as BT have ended their contract with ECI which means no more firmware support.  Also we dont know if the chipset in the ECI devices is as new as in the t-link's.

Regarding the dslam's it may be something similar, often hardware vendors will use software features to sell new hardware, so it could be ECI have told BT the M41 will only do downstream g.inp, if you want upstream support buy newer kit.  Or it could actually be a hardware limitation.

To me downstream only g.inp is not a huge deal as issues on the upstream are pretty uncommon compared to downstream.  Right now we have no g.inp at all, thats the bigger issue, and its not coming in 2015 either.

[NewtronStar]

To me downstream only g.inp is not a huge deal as issues on the upstream are pretty uncommon compared to downstream.  Right now we have no g.inp at all, thats the bigger issue, and its not coming in 2015 either.

You put that to AlecR and you will find it's rare but not uncommon for issues on the UPstream and the US G.INP needs to up and running in some cases.

[les-70]

  I have an intermittent upstream error issue which appeared to be roughly timed to the street lighting but came and went over about 4 months and seemed to be temperature dependent.  Not seen since late March but while it was present it would have given interleaving straight away.  I powered down the modem at dusk and for couple of months to keep things as they were.  I am hoping it does not come back when it get cold this winter.

   I would most certainly be very annoyed not see G.inp on up upstream. 

[burakkucat]

VTU-C64P V3 card

Interesting that you would appear to have documentation specifically discussing the linecard model.   
We know for a fact that Openreach are already using VTU-C64's.    V3 could possibly relate to the Vinax V3 chipset that is on those linecards  - we know that the existing linecards are capable of vectoring at line card level. 

The missing link for shelf based vectoring is and always has been space on the M41's backplane for a vectoring engine card.

Our special agent (codename Wheelbarrow) was active "in the field" when some ECI based cabinets were installed in Ewhurst, Surrey, a few years ago. Being a sociable person and able to establish a rapport with installers, some interesting images were thus obtained.

Example below . . .