$200 Asus Router ac1900+. which one? AC3100, AC3200, AC1900

[mogulman]

Coming from an RT-AC66. Wanting something from ASUS that has adaptive qos and ac1900 or higher.

Found the RT-ac3200 and the best buy AC1900p for both around $199.

Any idea which is better? I don't have a ton of wireless clients. Really just looking for range and wired wan and lan performance

 

[mogulman]

Coming from an RT-AC66. Wanting something from ASUS that has adaptive qos and ac1900 or higher.

Found the RT-ac3200 and the best buy AC1900p for both around $199.

Any idea which is better? I don't have a ton of wireless clients. Really just looking for range and wired wan and lan performance

Seems like the AC1900p has a faster CPU. Not sure if there is anything else better.

AC3200 is Tri--band, but not sure I care. Maybe AC3200 is just better anyway? It has more antennas.

 

[L&LD]

Between those two routers, I would prefer the RT-AC1900P (at least to test in my environment first with).

Simply because it is newer (design) and also/mostly because it will run RMerlin's firmware (choose the RT-AC68U download) too.

The RT-AC3200's hardware has been displaced by the latest generation of Asus routers. I would not be looking at that model as an option anymore.

In the RT-AC3200's case, more antennae are not better (just has an extra 5GHz radio). In some situations, a true 4x 4:4 antennae/stream configuration (or better) will lead to better performance even for clients with less than 4 antennae/streams, but this isn't the case with the Tri Band routers today which are 3x 3:3 max.

As your ISP speeds get higher and higher (I'm at 105d/23u and I feel I 'need' more than the RT-AC68U is giving me now as the first core is pegged at 100% and the second at 40% during speedtests), the need for a higher performing router (and processor) is more evident.

The RT-AC3200 is based on the same hardware as the original RT-AC68U (dual core, 800MHz processor). For the same cost ($199), the RT-AC1900P is the better router, imo, today.

 

[mogulman]

I guess a 3rd option in this price range is a Newegg Refurb RT-AC3100...$189.

So AC3200 new
AC1900p new
AC3100 Refurb

Anyone know if the RT-AC1900p has the Adaptive QoS?

 

[L&LD]

I guess a 3rd option in this price range is a Newegg Refurb RT-AC3100...$189.

So AC3200 new
AC1900p new
AC3100 Refurb

Anyone know if the RT-AC1900p has the Adaptive QoS?

It's been mentioned that the RT-AC1900P is otherwise an RT-AC68U with a better processor, so it should have Adaptive QoS.

I am not a fan of refurb's, but if you can test the RT-AC3100 fully with a full return policy, it is the better option, imo (latest circuit design, 1.4GHz dual core processor and 512MB of ram; double what the other solutions offer).

So, do you want to bet on a refurb?

 

[garyd9]

Just to throw another opinion in the mix:

Just because something is newer doesn't mean it's better. In many cases, something that's tried and tested is much better than something "newer." From what I can tell from these forums, this is especially true of Asus devices where the firmware for "newer" devices seems to need several months to iron out the issues. (It's bad enough that @thiggins comments on it in several of his device reviews.)

Consider that router technology is several years ahead of client technology, so that 3x3 device (AC1900) is actually MORE than most current wifi clients can use. There aren't many computers with 3x3 wifi devices, and NONE with 4x4. I just shopped for a notebook computer for my wife, and I wasn't able to find a single "mainstream" notebook with a 3x3 wifi device. They're all 1x1 or 2x2. (No, you can't just upgrade from 2x2 to 3x3 in a notebook if it doesn't have the antennas for it.)

It's a similar situation with all the various technologies that are being put into the latest router generations... they only exist in the router and no one knows if those techs will ever take off in client devices.

As far as processor speed, it's easy to buy into the marketing hype, but does it matter? Processing speed isn't going to make slow wifi go faster. It's not going to make ethernet switching faster. Sure, it sounds good on paper, but unless you're using your router for something other than routing and switching, the extra processing power is going to waste. It'd be like driving a car: You can only go as fast as the car in front of you.

Of course, you asked "which one is better?" That's subjective... the 1900p is newer and has a faster (newer) processor, but the AC3200 older, has more wifi radios and the firmware is more mature. For actual wifi performance, they'll likely be so similar that you wouldn't be able to tell them apart. The same for LAN performance.

I just went from a netgear R7000 (which is one of the highest rated AC1900 routers) to a Asus AC3200 (after hitting bugs that preventing me from keeping the Asus AC3100 and AC5300.) I have about 25 clients attached to the router at any given time (about 12 wireless and the rest wired via a switch) and I found that they are all about the same. No significant difference. I don't think I ever see the CPU usage go over "barely used." (Of course, I'm using it only as a router and not as a general purpose machine.)

No matter how you end up choosing, however, I wish you luck.

Take care
Gary

 

[TonyH]

Still most of us drive cars with puch Horse power including myself. Not much Hp is needed for moving people from point A to B, LOL! It is very true technology is advancing way ahead of us. This is so much wasteful at consumer level.

 

[mogulman]

Ok well. I can't tell too much difference between the Rt-ac1900 and the Rt-ac3200.

Physically the 3200 is bigger and has more antennas....,,,wooooooooooooo

Also there is Smart connect.

Speed wise they seem about the same.

The 1900 may have slightly better range.

Hmmm.....

 

[mogulman]

The 3200 must be better. Because its bigger!

 

[L&LD]

Ok well. I can't tell too much difference between the Rt-ac1900 and the Rt-ac3200.

Physically the 3200 is bigger and has more antennas....,,,wooooooooooooo

Also there is Smart connect.

Speed wise they seem about the same.

The 1900 may have slightly better range.

Hmmm.....

Can you tell us your test conditions? I hope you are not using 'Auto' channels? I hope you are using unique ssid's and not the same ssid's between both routers? I hope you tested for the best channel on each band for each router (they may or may not be the same, even between firmware versions).

 

[garyd9]

Can you tell us your test conditions?

I'm curious... do you expect there to be any noticeable difference between the two (outside of lab conditions)?

 

[mogulman]

Not auto channels. Not using both routers at the same time. Single SSID shared between both bands on a router.

I didn't do any scientific testing other than a few bandwidth tests and some tests for bufferbloat, which is the main reason I'm trying different routers. That and QOS that doesn't cut my speed in half.

The range remark is totally unscientific. Just me using my Surface 3 in different parts of my house. A couple of areas where an old Linksys would 't work. The AC66 worked much better. The AC1900 and AC3200 seemed to reach farther. Also unscientifically the Surface WiFi seems a little faster at further ranges. Then again, I'm probably trying to justify buying a router for the bufferbloat and qos. ☺

 

[L&LD]

I'm curious... do you expect there to be any noticeable difference between the two (outside of lab conditions)?

Yes. All the settings (and more) would make for very different results and therefore, not comparable in the end.

 

[garyd9]

Yes. All the settings (and more) would make for very different results and therefore, not comparable in the end.

Sorry, I meant to ask if you expected any noticeable difference between the two routers (outside of lab conditions)? (Perhaps I should say "significant noticeable difference given optimal configurations on each router")

If anything, for just wireless testing, I'd think that the 3200 _might_ have a slight advantage IFF the firmware can optimize which antenna(s) are used for any given connection. So, for 2 stream connections, it would be able to choose the "best 2" (out of 6) antennas for the specific client, while the 1900 would only be able to select 2 from 3. (I have no clue if the Asus firmware "dedicates" 3 antennas for 5-1 and 3 others for 5-2 or if picks them on the fly.)

(Now I'm getting into things which would be influenced by the angle of the antennas, if there was any metal nearby, reflections, atmospheric conditions, and other things that'd be extremely difficult to control.)

Of course, if I'm going to bring that stuff up, I'd be remiss in not mentioning that the antennas on the 1900p would be mounted higher in the air than the antennas on the 3200, given both routers sitting on the same surface.

(See why the best test is to just try them out?)

 

[TonyH]

The 3200 must be better. Because its bigger!

For me heavier is better, always, LOL!

 

[L&LD]

Sorry, I meant to ask if you expected any noticeable difference between the two routers (outside of lab conditions)? (Perhaps I should say "significant noticeable difference given optimal configurations on each router")

Yes to that too.

http://www.snbforums.com/threads/bufferbloat-and-qos-asus-ac66-merlin.33389/page-2#post-270133


The RT-AC1900P seems to give similar ping results with Adaptive QoS on as the RT-AC66U without it.

As the link above indicates, it may be a 'time of day' issue. But I think the much higher clocked processor with double the cores has something to do with it too.

 

[sfx2000]

Truly jealous of all you folks with 100Mbit connections out there - and bufferbloat isn't really an issue - at the house - I'm living with 50/5, and with a bit of tuning...

Slow bus...

 

[mogulman]

Interesting thing is a coworker of mine has At&t fiber 1 gb both directions and an RT-AC56. QoS turned off. He has no Bufferbloat. AT&T must be doing something on their end.

 

[sfx2000]

QoS turned off. He has no Bufferbloat. AT&T must be doing something on their end.

Again, Bufferbloat isn't really an issue once one is about 100Mbit/Sec on the WAN side...

ATT is probably doing a bit of work here as well - they've got some pretty smart networking folks...

 

[Lola]

I have Bufferbloat according to dslreports.com. However I don't have the slightest idea what buffrbloat is! Any one care to explain it?