Is there a policy at Mikrotik regarding product lifecycle/longevity?

We were building products based on the RB333, then it was EOL without much notice.

We were building products based on the Crossroads board, then without any notice it was EOL.

We were building products based on RB411A, then without any notice it was EOL.

We are now thinking about building a new product based on the RB411U, but want to make sure that it will be available for at least a year or two before EOL. We spend a lot of time and energy testing and qualifying products before we ship them. What might seem like a trivial change (RB411A to RB411U adds USB) to Mikrotik could be a support issue for us (how to handle RMA replacement for example), or may even be worse if some reliability issue, thermal issue, etc is introduced with a "trivial" change.

I know that technology constantly changes and that processors may come and go quickly, but it would be very helpful for those of us that are integrating and testing complete systems to have product stability and a published roadmap that anticipates extended lifecycles.

I worked at Intel for a while and we had an "Embedded Roadmap" for processors that Intel committed to keep available for at least 3 years so that board manufacturers and downstream integrators could be assured of availability. I would suggest that Mikrotik do a couple of things for us:

Create a long term product roadmap.
Commit to making certain products available (say RB433AH, RB411U, R52H, R52N) available for at least 3 years from first product shipped (might take some support from Atheros to get to this).
Have a clearly stated roadmap for replacement of EOL products (the RB411AR is not a comparable replacement for Crossroads for example because tx power is much lower).

Comments on this? Am I alone in this?

We are now using the RB4XX platform for all new products. They have all the same mounting holes and CPU architecture. Even if you can't get a RB411A, you can still get a similar product that will fit your boxes, and take the same configuration.

What might seem like a trivial change (RB411A to RB411U adds USB)

and removes some RAM.
Might need the RB411AH instead.

Is there a policy at Mikrotik regarding product lifecycle/longevity?

--cut--
Create a long term product roadmap.
Commit to making certain products available (say RB433AH, RB411U, R52H, R52N) available for at least 3 years from first product shipped (might take some support from Atheros to get to this).
Have a clearly stated roadmap for replacement of EOL products (the RB411AR is not a comparable replacement for Crossroads for example because tx power is much lower).

Comments on this? Am I alone in this?

I agree and think it would be great if Mikrotik would improve their communications on product cycle but I don't see it happening.

It would be nice if they would at least offer one last production run when they announce EOL. That way, OEM customers could buy up enough inventory to go another year or so. Instead, they usually announce EOL when they are about out of stock. My guess is that it leaves them less vulnerable in getting stuck with old stock.

What I have noticed since about 2005 when the RB532 came out and I started paying attention:

MT products have a short life span and are EOLed for 2 reasons.

1) They don't sell many/don't make enough money on the line.

2) Technology changes have made a better/cheaper product possible

Crossroads was a victim of #1. It wasn't a big seller and it became an orphan as the RB1xx family ended. There were other options like RB411A with radio card.

RB333 was a victim of #2. They realized the power of the Atheros MIPS chipset on the RB411 about the same time the RB333 was released. It allowed the RB433 at $99 when the RB333 was $159.

I think you will be safe if you ask your distributor what the best MT sellers are and go with that.

Tom

roc-noc do you think RB411U has a future?

I know the question was not directed at me, but in my opinion the RB411U is the wrong product.

They should have made an RB450/493 type device with the USB/3G this would suit business customers much more as they generally need more than 1 ethernet port.

Even better would be a new 7 series products e.g. the RB790GRU (Gigabit, Radio, USB)

- Attractive, professional looking case (with provision for micro-SD insertion)
- 9 Switch capable ports
- 1 WAN port
- 1x 802.11N wireless card
- 2x spaced USB ports (e.g. enough of a gap to allow a 3G modem and a USB flash drive to be inserted simultaneously)

This would provide a product that can be deployed to branch offices, small offices or homes. There are enough ports to connect PC's in, USB slots to allow a 3G connection and a micro-SD for proxy/logging. It could even be called the RouterBox

Future proof isn't really possible in this game.

However, Miktotickles should take note of the comments being made here.

Ubiquiti has recently tried a 'fek'em we know best' approach, and will be finding that their market share has dropped rather than what they wanted.

We like Mikrotik, and will probably keep buying your products. Your products are generally good.

However, we Can certainly make solutions with other manufacturers.

Piss off the punters through simple lack of communications, and you risk that.

Please just be polite and keep us informed about what you are selling us.

and a micro-SD for proxy/logging.

Anybody actually plugged a Micro-SD in and got it working ?

I'm on my 5th now, and none of 'em work.

I suspect that Mikrotik do not publish Road Maps for a reason. There are quite a few companies who view Mikrotik as their main competitor, and as soon as Mikrotik add a feature to their road map their competitors will try as hard as they can to beat them to market with it.

e.g.
NStreme is the reason AirMax exists.
RouterBoard is the reason RouterStation exists.

I think as far as Road Maps are concerned Mikrotik's best bet is to continue what they are doing.

As far as product development, maybe they need a few trusted partners they can run product ideas and prototypes by to make sure they are making a product the market wants.

Mikrotik's current model has brought them this far, so it is obviously working and working well, all companies have room for improvement but from what I have experienced Mikrotik are doing a really good job.

My main point is one of confidence in product availability and longevity. So when a board comes out MT could say "standard" or "extended lifecycle" on a part so that you can choose which product to use in which case.

Sometimes just having the same mounting holes, etc is not sufficient. We bid government contracts where deviations may or may not be allowed in the spec. Also, the integrator needs to step up to 3-5 years of supply. Without MT committing to something like this, we have to use other parts, often inferior to MT, because they are long lifecycle. You could even charge a little more for this feature.

Read about this at http://tinyurl.com/yddj5ap which is what Intel does. They have a subset of products that they commit to have available for 7 years from launch. While 7 years is kind of long, 2 or 3 would be great. This would let us all plan and expand the adoption of MT into more applications.

My main point is one of confidence in product availability and longevity. So when a board comes out MT could say "standard" or "extended lifecycle" on a part so that you can choose which product to use in which case.

Sometimes just having the same mounting holes, etc is not sufficient. We bid government contracts where deviations may or may not be allowed in the spec. Also, the integrator needs to step up to 3-5 years of supply. Without MT committing to something like this, we have to use other parts, often inferior to MT, because they are long lifecycle. You could even charge a little more for this feature.

Read about this at http://tinyurl.com/yddj5ap which is what Intel does. They have a subset of products that they commit to have available for 7 years from launch. While 7 years is kind of long, 2 or 3 would be great. This would let us all plan and expand the adoption of MT into more applications.

I second this, we finally were able to squeeze a rather ubiquitous statement in instead of giving a specific board, as when one of our government contracts started the 532 was standard, and we migrated through the 333 and now to the 433AH.

Our biggest issue has been the migration to 24v systems from 48v, which is the standard we deal with everywhere. We now have lots of POE switches doing nothing and a wall full of injectors. Nothing like taking a BIG step backwards.

Hi, RB532 and RB333 is still available!
http://www.routerboard.com/clearance.php

roc-noc do you think RB411U has a future?

We are not seeing very high sales of the RB411U yet but that should change now that we just sold out of the EOL RB411A last week. RB411U is the best fit replacement.

I think sales will also pick up as 3G service costs come down in the US.

The RB411U has two problems that I don't like compared to the RB411A. It only has 32MB RAM instead of 64MB like the RB411A. Also the high miniPCI socket (to clear the miniPCIe slot) doesn't allow you to use a radio card with MMCX and close the cover on the CA411 indoor case.

Tom

and a micro-SD for proxy/logging.

Anybody actually plugged a Micro-SD in and got it working ?

I'm on my 5th now, and none of 'em work.

I had Sandisk 2GB and 8GB both tested and working on the RB450G.

Tom

I'm using:

Sandisk 4GB class 2
Polaroid 8GB class 4
Polaroid 8GB class 6

and they all work perfctly.

FYI the Polaroids have identical production codes to Sandisk so save yourself some cash.

your not alone.. I invested approx $2000 in specific materials and months of time into a project that was based on crossroads.. just to see the effort and money go to waste since the crossroads hit EOL before the final tests were complete. crossroads was a great product just packaged wrong!

when I die I am going to will my wireless business to Normis as revenge LOL