On Mon, 2008-12-08 at 21:35 -0500, Tom DeReggi wrote: > I don't have a problem with you finding a reason or jsutification to use > Mikrotik. Mikrotik has a powerful unique product to save WISPs money. > However, I have a problem with your FUD misrepresenting facts. They are not > even close to accurate. You can't fairly compare apples to oranges either.
Which is the apple and which the orange. > Trango CPEs are $250-$300 per CPE and go for 12 miles, and never once in my > life had a Trango with a RSSI as low as -87. Under no shape or form will a > Mikrotik ever get a higher RSSI than a DSSS Trango radio of equivellent This is not necessarily true, either. The truth is that it depends on MANY factors. The radio itself is one of them. Mikrotik is not a "CPE", but an operating system (hence the name "RouterOS"). Just because you have not seen the lower RSSI values doesn't mean that it doesn't happen. > Mikrotik is LUCKY to get their stated full 23db of a Atheros HP card in > LOW modulations (not capable of 26mbps), and I've rarely seen it actually > deliver it consistently, after all the various potential places for loss > (UFl, pigtail, out of spec cards, lower grade filtering, etc). Hmm. Looks like you NAILED it! Using a superior operating system with quality components (good radio card and quality antennas/pigtails) CAN work as well as Trango or any other product on the market. > If your Mikrotiks are getting higher RSSI, then you are illegallly > over powering your Mikrotiks. This is a really broad statement and unfair accusation. You have no real idea if he is doing that or not. I don't know if he is or isn't, but the point is that neither do you. > You can't compare Trango's oldest product line to MIkrotik's newest. If you > are concerned about price you shouldn't be buying 5830s. There is a reason > that they made the FOX. You need to select the right product and buy savy > for Trango, just like you do for Mikrotik. You will also find that > Mikrotik doesn't do anywhere near 26mbps consistent throughput in a > scaled PtMP environment, expecially with the slightest amount of noise, > after combating all the congestion issues of a Wifi protocol (no > Nstreme polling does not perform as well as Trango polling). First, you are missing several realities of how MT works. Mikrotik's Nstreme is MUCH more than just polling. Nstreme offers 3 specific benefits, one of which is configurable for specific types of network traffic while the other 2 are simply a "switch". With Nstreme you now have the option to turn off CSMA. This fixes a LOT of the problem that outdoor wifi had in the first place. Then, you have polling. The polling mechanism has gotten a LOT of work recently. Versions after 3.15 (currently only in the test package) have a MUCH better polling mechanism and can scale very well. Perhaps not the the hundreds that a Canopy system can do, but then you don't need it to do that since you can build out more APs for the same $$. Finally, you have the other MAIN benefit of Nstreme, which is the packet aggregation feature. This feature is where the real benefit to Nstreme resides. You and I both know that typical IP traffic for most users is not even CLOSE to the 1500 byte MTU of Ethernet. The average packet size is MUCH smaller. Let's just say it's 200 bytes (this will vary a LOT, depending on the network). What the packet aggregation does is put multiple IP packets inside a single protocol frame. The policy that is used to determine if an IP packet goes into a frame that is being sent is configurable with 4 options. I won't go into detail on those options, as they are documented and you can go read about them. This aggregation technique can reduce the wireless network's overhead (thereby increasing timeslots available for real data) and can make a significant improvement in network throughput. It was back in 2004 that I upgraded a single point to point link and was amazed at the increase. This was a link that was running about 12Mbit throughput before Nstreme and simply turning Nstreme on for that link, it jumped to 18Mbit! That's a 50% increase (or 33, depending on how you calculate it). And that was WAY before the recent improvements in the protocol. The point here is this: You are comparing MT's polling to Trango's polling and the real benefit to MT's Nstreme isn't even in the polling mechanism. > There are many WISPs migrating to Mikrotik for some areas and applications. > But lets keep it real. Mikrotik has plenty of value, it is not necessary to > distort Trango's capability. I'd agree with this. But from looking at his post, it looked more like he was telling his own experience. Certainly he knows more about his own experience than you. Perhaps he is misreading the data, but that's not the assumption it looks like you are making. -- ******************************************************************** * Butch Evans * Professional Network Consultation* * http://www.butchevans.com/ * Network Engineering * * http://www.wispa.org/ * WISPA Board Member * * http://blog.butchevans.com/ * Wired or Wireless Networks * ******************************************************************** -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/
