@chojrak11: that sounds more like jobs for ciscos or junipers, because all these things especially the wire speed won't happen as fast as you probably need them in the mikrotik-world. maybe they put one rb with sfp-ports on the market, as thats nearly no work to do, but 10gbit on routerboards and asics will definitely take a while. mikrotik is focused on soho-markets and smaller isps.
I just daydreamed for a moment...
Seriously, I know it won't happen tomorrow, but we should say loud what we expect.
I moved my focus from Cisco to Mikrotik, as I service smaller ISPs, smaller companies and SOHOs, which can't afford Cisco, which anyway skins all its customers alive. I avoid anything but Mikrotik because it just works as expected, because its price is competitive, because it has outstanding feature set (which so far fits all needs of all my clients), and because I'm lazy, which means I don't want to learn any new crappy command line or management interface, or new software bugs. I know what to expect from MT and how to use it in a stable way.
The
bad thing (joke) is that these smaller ISPs grow, becoming not-so-small ISPs, so do their expectations for their core equipment. It would be a pity to throw away Mikrotik and put Cisco in this place, because it pops up obvious question "why we didn't buy Cisco in the first place but wasted money on this?"
Mikrotik has big potential and I'm sure they'll make new devices that will be more capable. I can see the evolution from 532 boards to where we are now. Their features are amazing, so no problem here. One thing I definitely wouldn't want Mikrotik people to do is support for exotic protocols like X.25, Frame Relay, exotic routing and so on. Leave that niche for Cisco. Do what's easy, already explored, and widely used. Which is what I posted earlier (ok skip ATM if you consider it exotic for now)
@hedele: I agree - CPUs in RB1000 should never be slower, no matter how many ASICs will be there!