I have looked through the MT & RB sites (and forums), and cannot find the pps/bps data for the RB433AH, and want to compare the comparative throughput.
Obviously the RB600 is quoted as the “Fastest RB ever built”, but this statement is now quite old, and it looks as if the RB433AH could give it quite a challenge! Is this why the specs haven’t been released? (i.e. MT want to sell all the stock of RB600’s first?)
The setup I am trying to achieve is for using NSTREAM2 for backhaul, as well as NSTREAM AP for our fixed CPE clients. I had chosen the 600 due to its PPS throughput (we offer a voip service), but am now considering the RB433AH due to the power/cost ratio ~ but really need to compare the throughput!!!
Has any USER done these tests?
Also, has anybody any idea what the new features are that should have been announced by now?
On one of the posts or data pages (I can’t remember where), there was a statement that the new RB433AH had 128Mb of memory that will be able to support “New Features” that will be announced in either spring or summer 2008 ( i.e. about now ).
So I was just wondering IF there was any chance of a “Heads up” on what these features may be, and if these alone would justify investment in the 433AH!?!
“The 128MB DDR will be capable of supporting new RouterOS features coming. The microSD slot supports an additional memory card that can be used for a Dude database and other features to be announced in during Spring ‘08.”
So this is what I was referring to, and the throughput comparisons still haven’t been updated. The current comparisons appear to show that the 433 is ‘slightly’ more powerful (on most tests) than the 333, but don’t give any indication of what the 433AH is capable of! Would it be possible to request that when they create these tests they publish BOTH the standard 680Mhz mode AND the overclocked 800Mhz mode ~ as I note a number of people have stability problems in the overclocked mode.
This would also be an interesting exercise to look at the difference between CPU vs bus speeds (i.e. If the bus can cope, the overclocking ‘should’ result in 15% more throughput ~ I realise this isn’t ‘that’ simple, but would be a good indication of if(?) / where any bottlenecks exist)
Surely the end of the statement “…features to be announced in during Spring ‘08.” ~ IS referring to SOMETHING specific!!!
Or is this just sales Bu115h1t?
If you’re stating that MT have NO ‘NEW’ features due to be announced, then that’s fine!!!
I must be just one of the muppets that your sales & marketing department target, expecting us to believe statements like “features to be announced” and “128MB DDR will be capable of supporting new RouterOS features coming” indicate that MT has a roadmap, and that the latest hardware MAY be needed to support features on it.
At least the “Duh” statement woke me up, and gave me a smile!!! How about YOU send an email to your sales & marketing department saying “Duh, what new features?”
Maybe, you could also ask them to post the updated throughput comparison tables as well!
OK, so if we want to compare for Nstreme Dual purposes, which will form > 1000 bytes packets, the tests are not much usefull. RB433AH seems to have 100mbit Ethernet ports, which went to its limit during tests, so the only comparison which applies is for 64bytes packets? According to that test, RB433AH seems to be some 30% faster than RB600 (bridge, conn-track, fw - off).
As RB1000 seems to provide much better results (but has 2 ugly fans , I wonder if it would not be better to look for some Atom based x86 small boards with passive cooling? Would make for Nestream Dual beast
I look the comparison but I think that RB600 is more powerfull than RB433AH , for big packet 1500 byte there is a lot of difference from RB433AH. I want to use for NSTREAM2 and backaul link where there is al lot of traffic .
Yes, RB600 has Gigabit ethernet ports, so it is not limited by that. Also it has 48V and Daughterboard support, and an additional miniPCI slot. So it has a lot of benefits too
Link please