Which Router Speedtouch 686v6 or Netgear DG834G v3

I now have 2 routers, I originally started off with the DG834G which in the evenings would drop to 0-1db SNR at points & maintain connection, the downstream attenuation would be about 53-54DB.

I recently purchased a ST 585v6 knowing that it may hold SNR better on my longer line, the stats I get on this in the evenings is 14-15DB SNR but my downstream attenuation is at about 58-59DB.

Which is best to use?! I am 4.5km from the exchange (thats not aa the crowflies but by cable length) i got the second router to try and increase my speed, current ip profile with bt is 750kbps.

On the netgear at points this has been at 1250 on the ip profile, however obviously now I am 500kbps less.

Realistically my speeds are the following :

[IMG]http://www.speedtest.net/result/40978989...[/IMG]

Also with the speedtouch i could try and alter my snr (i have the right firmware and DMT) but i dont want to [censored] my line up!)

Any help would be appreciated!

Regards



Ben

The 834v3 is based on the TI AR7 chipset and is a terrible router. Sure it holds 1dB margin but a better router like the 585v6 will achieve higher sync and the margin will not vary so much as it does with the 834.

The line attenuation is the same whatever the router but different routers may report it differently because there is no standard laid down for how it should be calculated (it is often reported as an average value across the frequencies used)

If you have any doors that need holding open then put your 834v3 to good use doing that. If not and you are totally devoid of morals you could put it on eBay but if you prefer to sleep soundly just bin it.

I use a Netgear DG834G v3 without problems (mind you I have a maximum sync).

What about the actual sync to the exchange then how can i stabalise it at possibly a higher speed? Just let the router run and see how BT assesses it using Adaptive Max Logic?

I have been running tests over the last year on dslzone which are saved in an overall speed history, in certain instances my line as been downloading at 900-950kbps (avergaing about 800) whereas at present I am struggling to get 500 in certain instances.

I know it doesnt sound a lot but when you havent got much in the first place 900 compared to 500 is an 80% increase.

I was on 6.5mb till i moved a couple of years ago to a nicer area and now i'm too far from the exchange!

Yes you are correct that they are fine on good lines and other aspects such as the user interface are better than the Speedtouch.

OK you can post up your full stats and I or somebody else will comment on what sort of speeds you may eventually expect.

What has happened is that the line has been unstable using the 834 and the BT system has imposed a higher target noise margin in an attempt to make it stable. That higher margin has reduced the sync you can achieve.

Regarding actual throughput it depends on how low the lowest recent resync has gone as to what IP profile the BTw system has set. It also depends on how good your ISP is.

Given stabiliy (which the 585 should bring) the IP profile will get corrected within days but it could take a month or so for your target noise margin to reduce to the default 6dB. It is possible to cheat a little on that using an early 585 firmware and a program called DMT which allows the default margin to be overridden.

One other thing worth asking is have you optimised your internal wiring by fitting a filtered faceplate, iPlate or removing the ringwire. It is best to run the router from the master socket too if this is at all possible.

The 585 is not a perfect router - the GUI is laughable and the wireless is weak - that isn't helped any by the fixed antenna either but in the area that matters on long line (the modem section) it is excellent. I personally found it outperformed the much hailed 2Wire 2700. The only place I found it beaten was by a Billion 7402 on UKOnline and that performs badly on O2 being completely outclassed by the 585v7 O2 box.

Later models of the 834 are fine routers. The v3 isn't which is probably why Netgear replaced the chipset on later ones.

Stats as of 13:28 today

Link Information

Uptime: 0 days, 10:02:01
Modulation: G.992.1 Annex A
Bandwidth (Up/Down) [kbps/kbps]: 160 / 864
Data Transferred (Sent/Received) [KB/MB]: 898.00 / 2.75
Output Power (Up/Down) [dBm]: 10.5 / 14.0
Line Attenuation (Up/Down) [dB]: 31.5 / 57.0
SN Margin (Up/Down) [dB]: 6.0 / 11.5
Vendor ID (Local/Remote): TMMB / 
Loss of Framing (Local/Remote): 4 / 0
Loss of Signal (Local/Remote): 1 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): 0
Error Seconds (Local/Remote): 6 / 0
FEC Errors (Up/Down): 0 / 97
CRC Errors (Up/Down): 0 / 2
HEC Errors (Up/Down): 0 / 2

I have a filtered faceplate and no ringwire is acting as an antenna to get interference on the line.

Regarding the DMT "cheat" I dont know how the slider for target SNR on the side bar measures the target snr as there seems to be no scale to it.

I have left interleaving enable on the line to try and correct any errors that I may be getting, could it be worth asking my ISP to force my ip profile to be reset through BT? Could this save waiting around.

In terms of the wireless of the 585v6, i was a little dubious at first as I had enahanced the standard antenna on my 834 to boost my signal, however my pc appears to be holding pretty solid on connection with it.

Oh I wish I lived closer to the exhange! Also the exchange is due for an upgrade to 21cn enabling up to 24mbps by the end of March. After trawling about on the net I dont think long distance lines have really seen much if any of an increase with this, and the main ones to benefit are the ones that live close to the exchange!

After looking at my stats on this page

http://www2.farina1.com/ADSL/default.asp...

It would suggest that I may be able to get a bit more out of my line but not much.
[censored] Broadband.

Ben

Well your sync looked abnormally low to me even given the elevated margin but I've noticed that the o/p power of the DSLAM appers to be lower than it would be on a better line which would account for that. I don't know why that is or if it will rise back to a more standard sort of level. It could simply be that your connection is using so few tones because of the low sync and syncing higher will cause the reported o/p to rise.

To tweak with the 585 you need 6.1.4.3 or lower firmware. That may still be on the Thomson site but if not I put it up on http://rapidshare.com/files/9812892/ST58... - you'll need to wait a short time for a free download.

I don't tweak now but from memory each notch on the scale represents 0.5db reduction of the target and you can pull it down from the default position a total of 10 notches or 5dB. If somebody currently tweaking could confirm or refute that it would be handy but if you only shift it down a few notches at first you can't do much harm. As I recall the procedure to actually do it was a tad obscure but once again from memory you had to set the slider then save and resync - without the correct firmware it doesn't show the tool so I can't use that to jog my memory. To use it at all you need to check unlock experimental area under special/options

Good luck.

As others have said that sync (even for your line attenuation/target SNR) looks very poor.

I know you are using a filtered faceplate but do your stats change is you plug directly into the TEST socket behind the removable faceplate? Also when you remove the face plate, which disconnects the internal wiring, it's worth checking all your phone equipment is dead in case someone has fitted a dodgy extension behind the Master Socket! House alarm fitters used to often do this...

NB. I'm another fan of the 585v6 and it works well on my line (just wish it had a better GUI as others have mentioned)

The markings on the DMT slider are at 10% intervals. But that isn't 10% of the numeric margin, as decibels are a logarithmic scale not linear.

You can reduce a 12dB target to 6.6dB I think, and a 9dB target to something lower, I forget what. A 6dB target can be reduced to somwhere between 1 and 2dB IIRC.

To put the tweak into effect it is the Apply/Resync button in the slider panel. Some routers allow the tweak to be saved but not SpeedTouch I don't think.