Why the difference in performance of N USB adapters

[Sei]

This is a thread more to clear my doubts on the topic rather than asking for a solution.

Scenario - I've got an Asus RT-N12 SuperSpeedN Wireless Router . This is kept close to a wall in my room.

In the adjacent room, which is the living room, I have placed my HTPC. The distance must be 14 feet, with a wall (separating both rooms) in between.

I had initially purchased an Asus USB-N13 USB stick for wireless streaming. Now, this setup performed really poorly. I used to barely get around 70-80% network strength. SD streaming was fine most of the time but 720p/1080p was out of the question.

I thought that this must be the case with wireless streaming but I had to purchase another wireless USB stick (ASUS stick was needed elsewhere) so I went ahead an bought a TP-LINK TL-WN822N USB stick.
The performance was mind-blowing!

SD/FHD streams without issues. I get a constant file copy speed of 6-7 MBps (tried with a 3GB ISO file). Even if a download is on in the background, I am easily able to stream media.

So, what exactly happened here? There has been no change in the configuration otherwise. Each and every component and location is the same. I checked the specs, but they seem the same to me.

Can anyone point out what I missed so that I am more aware in the future?

 

[blr_p]

Can anyone point out what I missed so that I am more aware in the future?

- Amplifier power output

Just look at the specs
ASus: Transmit Power : b mode: 18 dBm, g mode: 15 dBm, n mode: 15 dBm
Tp-link: Transmit Power : 20 dBm

3db means doubling of power, you have 5dbm difference here so it even more.

- Better antenna design means better receiver sensitivity as well.

Any model with the terms high gain or power uses a more powerful amp and better antenna design. This is more important than any dbm figures mentioned. Amped specialises in high gain adapters.

There are 3 versions of that tp-link adapter which one do you have ?

One way to improve performance with the N13 would be to use a USB extension cable (not more than 6 feet length). This way you can move the dongle about to where it gets a better reception. It won't beat the tp-link but it will be better than directly connected to the pc. Just noticed the tp-link already comes with an extension cable.

The same applies with routers as well. Vendors tend to cheap out on the amps used and some routers do not go as far as others.

What annoys me is they did not use proper amps in the dual band models for 5Ghz, but hey if its an ac model than everything you knew about 5ghz attenuation does not apply because the proper amps were used

Why don't you give us your inputs in this thread

http://www.techenclave.com/community/threads/movie-streaming-with-asus-rt-n13.146738/

That you can still get acceptable performance with an N12 is interesting.

 

[Sei]

- Amplifier power output

Just look at the specs
ASus: Transmit Power : b mode: 18 dBm, g mode: 15 dBm, n mode: 15 dBm
Tp-link: Transmit Power : 20 dBm

3db means doubling of power, you have 5dbm difference here so it even more.

- Better antenna design means better receiver sensitivity as well.

Any model with the terms high gain or power uses a more powerful amp and better antenna design. This is more important than any dbm figures mentioned. Amped specialises in high gain adapters.

There are 3 versions of that tp-link adapter which one do you have ?

One way to improve performance with the N13 would be to use a USB extension cable (not more than 6 feet length). This way you can move the dongle about to where it gets a better reception. It won't beat the tp-link but it will be better than directly connected to the pc. Just noticed the tp-link already comes with an extension cable.

The same applies with routers as well. Vendors tend to cheap out on the amps used and some routers do not go as far as others.

What annoys me is they did not use proper amps in the dual band models for 5Ghz, but hey if its an ac model than everything you knew about 5ghz attenuation does not apply because the proper amps were used

Why don't you give us your inputs in this thread

http://www.techenclave.com/community/threads/movie-streaming-with-asus-rt-n13.146738/

That you can still get acceptable performance with an N12 is interesting.

Aah, knew I was missing something. Makes sense since I saw a dramatic increase in network performance.

I'm using this --> http://www.tp-link.com/en/products/details/?model=TL-WN822N

I did have a USB extension cord for the ASUS N13 stick and pointed it towards the router. Didn't really help. The only difference right now is that the TP-Link sits at the same horizontal level as the router and the ASUS N13 stick was a bit below. I'm not sure if that matters.

I only use XBMC for viewing media. It might be that they have done some enhancements in the streaming part of their software?

Initially, I had suspected the router of the bad performance but it seems that it was the N13 stick all along. Any more tests you would like me to run?

 

[blr_p]

Aah, knew I was missing something. Makes sense since I saw a dramatic increase in network performance.

The difficult bit is figuring this out BEFORE you buy stuff. The only clue is 'high gain'.

I'm using this --> http://www.tp-link.com/en/products/details/?model=TL-WN822N

I know but there are three versions of that adapter. Read this

I want to know if its V1, V2 or V3.

What networking vendors do over time is change stuff under the hood for cost cutting purposes. This does not always mean the performance goes down but it can be important if alternative firmware is to be used. Course N/A in terms of your adapter but matters when its the router.

I did have a USB extension cord for the ASUS N13 stick and pointed it towards the router. Didn't really help. The only difference right now is that the TP-Link sits at the same horizontal level as the router and the ASUS N13 stick was a bit below. I'm not sure if that matters.

Not really. The TP-link is a better product for less, just goes to show how aggressive tp-link is about gaining market share that they will even forgo profits to get a good rep.

The bad news for most people is they cannot use this tp-link adapter with their smart tv's or media streamers. Those companies want to sell their not so high gain adapters at much higher prices. It will only work on platforms that allow you to install this tp-link driver

I only use XBMC for viewing media. It might be that they have done some enhancements in the streaming part of their software?

What matters is it use DLNA and if you can get full HD going i bet it does. i take it you are watching this on a PC, what viewing client are you using to watch the movies.

Initially, I had suspected the router of the bad performance but it seems that it was the N13 stick all along.

There's 3 elements at play here. Server - network - client.
At the physical level its router-adapter
At the network filesysteml level its SMB/DLNA
At the application level its the software the server uses and the client used for playback.

Any more tests you would like me to run?

Want to know the highest bitrate file you managed to stream across successfully. Pick a large filesize, full HD movie with lots of action.

If you media player has a media info option it will show something like 'overall bit rate'.

There is another tool you can use also, called bitrate viewer. This will show the avg as well as peak rate. When it runs ensure that the movie duration it reports is the same as the duration your media player shows.

When people say they managed to stream HD/Full HD or 720p/1080p files it does not tell us much. 720p files can have bitrates from 5Mbs-15Mbs. 1080p can be anywhere from 8-30Mbs. Depends on the encoding. So what matters is the bitrate. If your files are low bitrate then you got away. When the rates are higher, the peaks are even more and then the ability of router+adapter to stream consistently comes in.

 

[Sei]

The difficult bit is figuring this out BEFORE you buy stuff. The only clue is 'high gain'.


I know but there are three versions of that adapter. Read this

I want to know if its V1, V2 or V3.

What networking vendors do over time is change stuff under the hood for cost cutting purposes. This does not always mean the performance goes down but it can be important if alternative firmware is to be used. Course N/A in terms of your adapter but matters when its the router.


Not really. The TP-link is a better product for less, just goes to show how aggressive tp-link is about gaining market share that they will even forgo profits to get a good rep.

I will check that - wasn't aware of it.

The bad news for most people is they cannot use this tp-link adapter with their smart tv's or media streamers. Those companies want to sell their not so high gain adapters at much higher prices. It will only work on platforms that allow you to install this tp-link driver

What matters is it use DLNA and if you can get full HD going i bet it does. i take it you are watching this on a PC, what viewing client are you using to watch the movies.

There's 3 elements at play here. Server - network - client.
At the physical level its router-adapter
At the network filesysteml level its SMB/DLNA
At the application level its the software the server uses and the client used for playback.

That's true.

Using XBMC to watch movies and shows - www.xbmc.org

Want to know the highest bitrate file you managed to stream across successfully. Pick a large filesize, full HD movie with lots of action.

If you media player has a media info option it will show something like 'overall bit rate'.

There is another tool you can use also, called bitrate viewer. This will show the avg as well as peak rate. When it runs ensure that the movie duration it reports is the same as the duration your media player shows.

When people say they managed to stream HD/Full HD or 720p/1080p files it does not tell us much. 720p files can have bitrates from 5Mbs-15Mbs. 1080p can be anywhere from 8-30Mbs. Depends on the encoding. So what matters is the bitrate. If your files are low bitrate then you got away. When the rates are higher, the peaks are even more and then the ability of router+adapter to stream consistently comes in.

I understand what you mean. I simply tried playing back two 10GB 1080p files that were unplayable before. I will check using the bitrate viewer too and report back.

 

[blr_p]

Using XBMC to watch movies and shows - www.xbmc.org

Which platform is the XBMC client running on ?

As i think about that N13, i wonder if its drivers were just not that good. The N13 is a N300 adapter though from your description it felt more like an N150.
Did you remember what link speed you got with it.
At 20Mhz
65-72 - N150
130-150 - N300

At 40Mhz the figures double.

Are you running the N12 in 20Mhz or 40Mhz, did you notice any difference.

 

[Crazy_Eddy]

Any model with the terms high gain or power uses a more powerful amp and better antenna design. This is more important than any dbm figures mentioned. Amped specialises in high gain adapters.

IINM, transmit power figures in dbm include the antenna gain as well.

Another point to note - the USB-N13 has a Rev A1 and B1. The B1 has a useless Realtek chipset.

That you can still get acceptable performance with an N12 is interesting.

The N12 is a very under-rated router. Broadcom chipset + tomato firmware would make it an excellent buy. Unfortunately only the A1 version of this router was available in India, and there were some reports of instability with tomato on this version. The newer B1/C1 revisions were better, had 5dbi gain antennas and even have AsusWRT support.

 

[Sei]

Which platform is the XBMC client running on ?

Oh! Well, you did say "viewing client" earlier. Anyways, I'm running it on Windows 8.

As i think about that N13, i wonder if its drivers were just not that good. The N13 is a N300 adapter though from your description it felt more like an N150.
Did you remember what link speed you got with it.
At 20Mhz
65-72 - N150
130-150 - N300

At 40Mhz the figures double.

Are you running the N12 in 20Mhz or 40Mhz, did you notice any difference.

AFAIR, I was running it in 20Mhz. I will check and revert.

 

[blr_p]

IINM, transmit power figures in dbm include the antenna gain as well.

I have no idea where flipkart got those figures, the vendors site says EIRP <20dbm. There is no reliable source. If the vendor pushes it then they think they can charge more for the product.

Another point to note - the USB-N13 has a Rev A1 and B1. The B1 has a useless Realtek chipset.

Bad move, just spoils the name of the N13. It was strange, ok tp-link is better but THAT much better than the N13 is surprsing (!)

xtreamer has a realtek too and from what i can tell its a N150 even though the chipset is a 2x2.

The N12 is a very under-rated router. Broadcom chipset + tomato firmware would make it an excellent buy. Unfortunately only the A1 version of this router was available in India, and there were some reports of instability with tomato on this version. The newer B1/C1 revisions were better, had 5dbi gain antennas and even have AsusWRT support.

Yeah, this is an openwrt model.

Another point for Sei to confirm, which version of N12 does he have ? A,B,C or D.

 

[Crazy_Eddy]

the vendors site says EIRP <20dbm.

Yeah EIRP includes antenna gain.
Just noticed on WikiDevi, the V3 of the TP-Link has switched to the same Realtek chipset. Thankfully its a very recent variant, so probably hasn't hit India yet. Wonder what the performance difference will be like with both Realtek based dongles head to head!

Another point for Sei to confirm, which version of N12 does he have ? A,B,C or D.

The A1 was the only variant listed on the Asus India site, so I'm guessing this was the only variant launched in India. A friend picked one up recently and it was still the A1 variant. Hope @BIKeINSTEIN can confirm. I'm guessing a B1/C1 would've cannibalized the RT-N13 market. But Broadcom/tomato loyalists like me are left with no option between the N10 and the N16.

 

[blr_p]

Yeah EIRP includes antenna gain.
Just noticed on WikiDevi, the V3 of the TP-Link has switched to the same Realtek chipset. Thankfully its a very recent variant, so probably hasn't hit India yet. Wonder what the performance difference will be like with both Realtek based dongles head to head!

Hehe, so how wierd it would be if Sei came back and confirmed that a V3 what he had. Most likely its a V2.

So this is a message to the future to people reading this thread 6 months or 1 year after April 08 2013.

VERSIONS MATTER, JUST COS IT WORKS FOR SEI DOES NOT MEAN IT WILL FOR YOU !

You have been warned.

The A1 was the only variant listed on the Asus India site, so I'm guessing this was the only variant launched in India. A friend picked one up recently and it was still the A1 variant. Hope @BIKeINSTEIN can confirm. I'm guessing a B1/C1 would've cannibalized the RT-N13 market. But Broadcom/tomato loyalists like me are left with no option between the N10 and the N16.

Not seen bikey here in ages.

N12 does not have USB, so no 3G possibility, which makes the N13 untouchable.

What's so nice about tomato ? You have the same 1k+ apps availiable like the dd-wrt/openwrt people.

 

[Sei]

Wow, this has become an interesting discussion!

Alright, so here are the figures (and both @Crazy_Eddy and @blr_p are correct) -
TP Link Adapter- Version V2
ASUS N12 router - Version A (No version mentioned anywhere, not on the bottom or on main homepage, I am assuming A)
ASUS N13 USB stick - Seems to be Version A1. I checked at http://wikidevi.com/wiki/ASUS_USB-N13_rev_A1and looking at the H/W ID in Device Manager, I can see "USB\VID_0B05&PID_1784&REV_0101" and "USB\VID_0B05&PID_1784"

 

[blr_p]

ASUS N13 USB stick - Seems to be Version A1. I checked at http://wikidevi.com/wiki/ASUS_USB-N13_rev_A1and looking at the H/W ID in Device Manager, I can see "USB\VID_0B05&PID_1784&REV_0101" and "USB\VID_0B05&PID_1784"

If you have the original box, see here.

If its indeed an A1, then this is the second part of the wierd scenario.

He's got a ralink in there yet it underperforms

Come to think of it we never did get to see a transfer speed test, would it be possible to get one with the N13 usb ?

SD/FHD streams without issues. I get a constant file copy speed of 6-7 MBps (tried with a 3GB ISO file)

How was this measured. Windows explorer gives a MBps read ?

 

[Sei]

If you have the original box, see here.

If its indeed an A1, then this is the second part of the wierd scenario.

He's got a ralink in there yet it underperforms

Come to think of it we never did get to see a transfer speed test, would it be possible to get one with the N13 usb ?


How was this measured. Windows explorer gives a MBps read ?

Didn't get the time to do much else except check the version numbers.

Yup, the new in-built copy tool of Windows 8 does - http://blogs.msdn.com/b/b8/archive/...ement-basics-copy-move-rename-and-delete.aspx

 

[blr_p]

ASUS N13 USB stick - Seems to be Version A1. I checked at http://wikidevi.com/wiki/ASUS_USB-N13_rev_A1and looking at the H/W ID in Device Manager, I can see "USB\VID_0B05&PID_1784&REV_0101" and "USB\VID_0B05&PID_1784"

Yes i think we can show that your USB-N13 is an A1

device id : 1784 is the A1 Ralink

device id: 17ab is the B1 Realtek

Why there is such a difference is unclear. Mis-configured i suspect. It would be reasonable to expect a 20-50% boost with the tp-link but not a 100% and above increase over the N13.

lets see what the ISO copy rate is like with this adapter.

Basically something is preventing the N13 from functioning as the N300 device it really is and instead its working as a N150 or even g (likely). One avenue to check is the encryption, any WEP setting will force g speeds. You have to set it to WPA2+AES to get the best performance. No TKIP.

Also set the router to n only, no mixed b,g,n etc. This will eliminate the possibility of a g device (if you have any) associating with the router just before the transfer is initiated. You can't share a video network with slow devices, it has to be exclusive or failing that, n only, to maintain your transfer speeds.

The tp-link is better with operating in a mixed environment than the N13 even though both are certified 'n'. Neither is a draft device.

Then try the ISO copy test and see if there is any differnce.

 

[blr_p]

I thought that this must be the case with wireless streaming but I had to purchase another wireless USB stick (ASUS stick was needed elsewhere) so I went ahead an bought a TP-LINK TL-WN822N USB stick.

Do you still have that N13 USB otherwise we'll not be able to understand why there was such a difference.

Still waiting to hear what the biggest size file size you managed to stream across ?

 

[Sei]

Do you still have that N13 USB otherwise we'll not be able to understand why there was such a difference.

Still waiting to hear what the biggest size file size you managed to stream across ?

Do you still have that N13 USB otherwise we'll not be able to understand why there was such a difference.

Still waiting to hear what the biggest size file size you managed to stream across ?

I'm sorry for the delay in response.. been distracted for the past few days, busy work and home stuff. I do have the USB N13 with me and I'll post results tomorrow for sure.