Print

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

[Weaver]

[This topic has been created by splitting off the following posts from g3uiss' "At Last" thread.]

Congratulations!

Do you think that router manufacturers are cautious in their maximum throughput figures?

[Alex Atkin UK]

Congratulations!

Do you think that router manufacturers are cautious in their maximum throughput figures?

I suspect the answer is no, that's the NAT throughput without PPPoE which will drag down the speed even more.  I'd be happy to be proven wrong, but we shall see.

In 99.9% of cases manufacturers don't even tell you the routing speed and most people reviewing routers seem to not understand how they work so don't test it either, they just plug it into their broadband and say "I get full speed" which is really not helpful if your broadband is faster than theirs.

[j0hn]

Do you think that router manufacturers are cautious in their maximum throughput figures?

Usually the opposite.
Many quote the throughput achievable with firewall disabled and no PPP.

Draytek being 1 of the few who will quote the firewall throughput.

[craigski]

I suspect the answer is no, that's the NAT throughput without PPPoE which will drag down the speed even more.  I'd be happy to be proven wrong, but we shall see.

I'm curious, can you elaborate regarding dragging speed down using PPPoE. I'm assuming if an ISP quotes a speed and they provide the service using PPPoE, then the speed they quote will include any overheads of their network/equipment?

[j0hn]

I'm curious, can you elaborate regarding dragging speed down using PPPoE. I'm assuming if an ISP quotes a speed and they provide the service using PPPoE, then the speed they quote will include any overheads of their network/equipment?

I believe he's discussing consumer hardware that struggles with PPP rather than any PPP packet overheads.

ISP's quote sync speeds which take no account of PPP.
The packet overhead added by PPP is also absolutely tiny (8 bytes)

[craigski]

I believe he's discussing consumer hardware that struggles with PPP rather than any PPP packet overheads.

ISP's quote sync speeds which take no account of PPP.
The packet overhead added by PPP is also absolutely tiny (8 bytes)

I guess what I am trying to say is if an ISP offers a '500 Mb/s' service, they provide this service over PPPoE, and they provide the router/firewall, the ISP will have tested this NAT/PPPoE configuration can support up to 500Mb/s quoted and the 500Mb/s includes any overheads network in their network hardware/protocols to provide this service.

[d2d4j]

Hi

I would consider they do, so ISP provided router would give advertised speed as tested from router to ISP termination point.  However, the firewall would be basic and in supplied mode, only contain a couple of rules.  This to me though is similar to the connected speed of ADSL/FTTC as that is not the speed you would have but the connected speed.

However, please note the minimum guaranteed speeds - using 500/70 package from BT

Average
501
Mbps
Download
Average
68
Mbps
Upload
Minimum guaranteed speed: 250 Mbps download

Also, you would need good equipment/computer to use that speed, anything low powered would not be able to handle the throughput @500mb/s

The Draytek G3uiss posted (2927) is rated performance at 1000 mb/s at firewall I believe.

I do not understand why you are looking for exact speed at those speed throughputs... do you complain over your car consumption to advertised car consumption or your fridge or washer using exact power as advertised...

Also, I guess not many realise but John posted that FTTP is not as reliable as FTTC and I concur, it is not (see picture wan 1 BT FTTP 900/900 wan 2 FTTC 80/20).  I also noticed recently that BT state 48 Hours to fix and I cannot see the 900/900 package anymore

48 hour resolution
We will aim to fix your fault by the end of the second working day after you’ve told us about the fault.

Please do not think I am having a go, it's just something I do not understand with people so not trying to upset anyone

Many thanks

John

[g3uiss]

Well said @d2d4j The 2927 is rated at 1.8gb firewall thruput and with some very comprehensive options. VPN is less but quite adequate I think.

As I’m now essentially a domestic user+ I’m not having a FTTC fall back, but just a 4G on VOXI.

Tony

[Alex Atkin UK]

Yep, ax supports USB (WAN5/6), but not wireless WAN (WAN3/4).

Wireless WAN is something different, you use the wireless adapter in the Draytek to tether via WiFi to a hotspot (eg phone/mifi device) for backup connectivity:

https://www.draytek.co.uk/information/our-technology/wireless-wan-wifi

I'm impressed any of their models support it, I've only ever seen it doable on DD-WRT/OpenWRT, not on stock firmware.  Although mobile routers with hardwired ethernet ports usually also have better reception/speeds, so my backups are both hard wired with the WiFi left on so I can use them to test remoting into the network over the DSL line by connecting a laptop directly to the mobile router.

[craigski]

Wifi WAN - not an option for our clients given the bandwidth usage and forgive me, but is that not the same as 4G dongle, which may work better then wifi wan as it is direct to router rather then making a wifi connection to hotspot and mobile or other device then making WAN connection.  As you know, we have 4G dongles in use on drayteks which work lovely.

so my backups are both hard wired with the WiFi left on so I can use them to test remoting into the network over the DSL line by connecting a laptop directly to the mobile router.

Can you recommend a 5G capable device that I can use for backup of FTTP on Draytek? I used to have 2x FTTC with different ISPs, so looking to replace those with something faster for FTTP redundancy. I'm struggling to find mobile operators that will offer a 5G USB solution, as most tend to be WiFi with Ethernet now. As my 2927ax wont support WiFi WAN, Ethernet is preferred, as it will allow me to locate device away from router for better reception etc.

TIA!

[Weaver]

Craigski, could you use a WAP in ‘client mode’ and link that to one of these wireless LAN+devices? I know the latter might not be ideal because they could do NAT, act as routers rather than simple bridges. I’m thinking of using AA’s L2TP service to get round the problem of wireless devices being ‘rich’ routers.

[Alex Atkin UK]

I've been really pleased with the Hauwei CPE Pro 2 (H122-373) although its now £100 more expensive than when I bought it new unlocked on Amazon. :/

I've just ordered a used Three locked one from CEX so I can move my unlocked one to Lebara as even on 4G it got 20Mbit more than the B353 and it means should they ever roll out 5G I don't need to keep checking.

[d2d4j]

Hi

I am sorry, we do not use 5G at this moment and my experience of 5G is not good for speed, so Alex Atkin Uk looks the better option

We are surrounded by 5G and here is our speed as taken from an Iphone 12 on latest IOS.  4G is a lot faster

Many thanks

John

[craigski]

Thanks all, some food for thought, I will let you know how I get on. And sorry to OP for hijacking thread going a bit off topic.

[meritez]

I suspect the answer is no, that's the NAT throughput without PPPoE which will drag down the speed even more.  I'd be happy to be proven wrong, but we shall see.

In 99.9% of cases manufacturers don't even tell you the routing speed and most people reviewing routers seem to not understand how they work so don't test it either, they just plug it into their broadband and say "I get full speed" which is really not helpful if your broadband is faster than theirs.

Mikrotik give the routing speed for all of their products:
https://mikrotik.com/product/ltap_mini#fndtn-testresults