• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

giga routers total throughput ?

Elixer

Elixer

Lifer
Currently, got the die-hard wrt54G, and it just can't handle the speeds anymore on the WAN side.
It currently is doing 100% load (DDR-wrt firmware) when downloading, and once that happens, then it gets pretty ugly fast, for anything else trying to connect via wireless.
The wrt54G is suppose to handle 100mbps, but it goes 100% load at around 26 mbps on any port, and then it just gets ugly, so got a nice giga switch to get around that issue for the LAN side.

Trying to consolidate into one device, was looking at http://www.newegg.com/Product/Produc...33162069-L012D
And while it says it is a giga router, I am wondering just how much throughput can this handle ? Yeah, in theory, 1000mbps, but is that total throughput for all the ports on it, or is it, each port (WAN included) can handle full gigabit connections ?
 
Maximum router throughput will vary based on packet size, firewall functionality, simultaneous connections, etc., so a "total throughput" figure wouldn't make much sense. You can derive a best-case total throughput by multiplying the router's maximum packets per second by the size of the packets it's sending, but consumer router manufacturers rarely publish PPS numbers.

Your best impartial resource for consumer/SOHO networking equipment is SmallNetBuilder, and they have a review of the router you're interested in. They claim that total throughput is about 180Mb/s, and while I haven't looked at their testing methodology, that's a reasonable ballpark figure for a modern SOHO router.
 
I used the WZR-300HP for a while. On the LAN (switch side) it perform like any Giga switch.

On the WAN to LAN side I never tested it. I do not know the limit since my Internet is only 20Mb/sec. (with temp. spikes of 32Mb/sec.), the WAN side performed with such bandwidth well.

P.S. At the first 5 years of market availability the specs of the entry level Routers where not functional as stated. No one at the time cared since the Internet bandwidth was low. Many of the simple Routers' Data sheets indicated 100 Mb/sec. on the WAN side but they really could handle such speed.

In general, even today the Marketers of Entry Level Wireless Routers allow themselves "Artistic creativity" in Data sheets, rather than adhering to real technology parameters and standards.


😎
 
Last edited:
Funny you mention data sheets, was trying to find some that actually have some solid information, instead of the marketing BS... no luck yet.
You call them up, and ask a few questions, and they either have no clue, or say buy it, try it, and return it if you don't like it.
 
Just a FYI, I obviously don't have a 100Mbs link according to the great support staff @ Cisco. Never mind I told them I am getting 26Mbps max.

Their lab tested the wrt54G, and they told me, their labs show both WAN & LAN can handle exactly 100Mbs.

I just have to laugh.

I asked for the whitepaper, just to see what they will tell me.
 
I'm curious as to the testing methodology as well. Are they using large packet sizes windowed TCP flows? are they using NAT? are any other "features" turned on. . . Elixer, please report back with that white paper cause I'm interested to see as well!
 
Yeah, seen that before.
This is the normal WRT-54G predates the GL version.

I still think this is false advertising, if you claim something goes X as fast, it should get pretty close to that X, not X/4.
 
Have you tried running tomato? I was getting 20-30 mbit on my WRT54GL running DD-WRT(50mbit connection); tomato got it much closer to 50 mbit. I decided at that point to just dedicate an atom pc to router duty and have it running pfense currently, but the WRT54GL came pretty close running Tomato, much better in that regard than DD-WRT.(though overall I still prefer DD-WRT)
 
Yeah, tried all available firmwares.
The only thing you can do is o/c the CPU on this wrt54G, from 200MHz to 228 (any more, and it crashes), and that gets it up to 29Mbps.
 
The WRT54G/GL has a slow processor and pathetic WAN to LAN throughput, compared to a modern "N" router.

Edit: I am currently using E2500 Cisco routers, with Shibby Tomato 1.08 installed. I have 75/35 FIOS (84/39 in practice), and I get full speed with a wired connection through the E2500. Even though it lacks Gigabit ports.
 
Last edited:
VirtualLarry said:
The WRT54G/GL has a slow processor and pathetic WAN to LAN throughput, compared to a modern "N" router.

Edit: I am currently using E2500 Cisco routers, with Shibby Tomato 1.08 installed. I have 75/35 FIOS (84/39 in practice), and I get full speed with a wired connection through the E2500. Even though it lacks Gigabit ports.
Click to expand...

Yeah, I know this, but Cisco/linksys says it can support "exactly 100Mbps".
I say BS.
Waiting for them to respond again.
 
Elixer said:
Yeah, I know this, but Cisco/linksys says it can support "exactly 100Mbps".
I say BS.
Waiting for them to respond again.
Click to expand...

They may be right under a carefully constructed scenario (and they are right that 100Mb/s per interface is an absolute upper limit), but the router is clearly not able to hit 100Mb/s in a real-world situation, and I doubt that Cisco/Linksys ever guaranteed that it would.

Since it sounds like you're aware of the WRT54G's performance limitations, I'm not sure what you're hoping to achieve by continually bugging Linksys about it, particularly considering that the model has long since been discontinued.
 
Most router and switches can handle more total throughput by 2-4 times of a single port, This is hard to test without having multiple sorce destination systems to test with simultaneously. (Not counting WAN)

Maybe Ive been reading it all wrong but wouldnt 54g limit be 54Mb/s which would be about 26Mb/s bidirectional on a single antenna (single band), And at best 43Mb/s multi antenna (duel band)?. (Not accounting for noise)
And then N is from 54Mbs - 400Mb/s.
 
theevilsharpie said:
They may be right under a carefully constructed scenario (and they are right that 100Mb/s per interface is an absolute upper limit), but the router is clearly not able to hit 100Mb/s in a real-world situation, and I doubt that Cisco/Linksys ever guaranteed that it would.

Since it sounds like you're aware of the WRT54G's performance limitations, I'm not sure what you're hoping to achieve by continually bugging Linksys about it, particularly considering that the model has long since been discontinued.
Click to expand...

I am pointing out that there is no possible way that the WRT54G can handle anywhere near the rated 100Mbps, even under any specific tests that would have been done. The Broadcom BCM4712 @ 200MHz just isn't fast enough to handle that amount of data. The hardware itself, may be able to handle it, but not the punny CPU that Cisco/linksys used.
I am hoping that this will lead to them releasing more information on how they test things.
Lorne said:
Most router and switches can handle more total throughput by 2-4 times of a single port, This is hard to test without having multiple sorce destination systems to test with simultaneously. (Not counting WAN)

Maybe Ive been reading it all wrong but wouldnt 54g limit be 54Mb/s which would be about 26Mb/s bidirectional on a single antenna (single band), And at best 43Mb/s multi antenna (duel band)?. (Not accounting for noise)
And then N is from 54Mbs - 400Mb/s.
Click to expand...

This has nothing to do with wireless.
This is wired.
 
There's not likely a consumer router that can "route" at wirespeed at 1000 Mbs. You need layer3 switching hardware to do that, not software.
 
Elixer said:
I am hoping that this will lead to them releasing more information on how they test things.
Click to expand...

Good luck with that.

It's a $60 router. The odds of them testing it beyond "does it boot?" are pretty low. Linksys may genuinely not know the maximum actual throughput of any of the routers that they sell, and even if they do, the throughput depends on so many variables that they would likely be unwilling to share their testing results.
 
Lorne said:
Maybe Ive been reading it all wrong but wouldnt 54g limit be 54Mb/s which would be about 26Mb/s bidirectional on a single antenna (single band),
Click to expand...

The 54g simply stood for the fact that the 802.11g (the "g" in 54g) is rated at 54Mbits per second theoretical speed (the "54" in 54g).
 
Elixer said:
Currently, got the die-hard wrt54G, and it just can't handle the speeds anymore on the WAN side.
It currently is doing 100% load (DDR-wrt firmware) when downloading, and once that happens, then it gets pretty ugly fast, for anything else trying to connect via wireless.
The wrt54G is suppose to handle 100mbps, but it goes 100% load at around 26 mbps on any port, and then it just gets ugly, so got a nice giga switch to get around that issue for the LAN side.

Trying to consolidate into one device, was looking at http://www.newegg.com/Product/Produc...33162069-L012D
And while it says it is a giga router, I am wondering just how much throughput can this handle ? Yeah, in theory, 1000mbps, but is that total throughput for all the ports on it, or is it, each port (WAN included) can handle full gigabit connections ?
Click to expand...
The behavior on your wrt54g is normal by the way. Even the stock firmware does this. It is physically capable of 100mbps but they never put in a processor fast enough to actually do so across the wan. I believe if you actually read the documentation on the orginal wrt54g, it mentions that wan throughput is limited to 25mbps.

You should be able to get more speed if you switch to tomato firmware. dd-wrt has been stagnant for years now. I switched to tomato because its usb support is SOO much faster. It's better to just bet a new router with a much more powerful processor.
 
Just as a comparison, on my internal network (rated for 1Gbps) i get a constant 112MBps transfer rate over the wired network as long as it's a single large file and not a batch of small files (otherwise the speed drops accordingly). On the wireless network i get between 8 and 12MBps depending on where i am in the house.

It's a new installation with a brand new switch and router with new cabling as well.

This router:
http://www.asus.no/Networks/Wireless_Routers/RTN66U/

This switch:
http://www.netgear.com/service-provider/products/switches/unmanaged-desktop-switches/gs108.aspx

Very nice stable setup.
 
smakme7757 said:
Just as a comparison, on my internal network (rated for 1Gbps) i get a constant 112MBps transfer rate over the wired network as long as it's a single large file and not a batch of small files (otherwise the speed drops accordingly). On the wireless network i get between 8 and 12MBps depending on where i am in the house.

It's a new installation with a brand new switch and router with new cabling as well.

This router:
http://www.asus.no/Networks/Wireless_Routers/RTN66U/

This switch:
http://www.netgear.com/service-provider/products/switches/unmanaged-desktop-switches/gs108.aspx

Very nice stable setup.
Click to expand...

Very nice router, but most people probably won't pay $170 for a new router when $30 options exist.

Are you getting 8-12 MBps on 5Ghz?
 
Elixer said:
Hmm, didn't know about these... was looking around, and it seems MikroTik RB751G-2HnD is pretty darn good spec wise, and price is under $80.
Click to expand...

they are not plug and play, but if you invest some time how to configure them, nothing beats them on performance, specs and price
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top