Ceee means 1 control channel plus 3 extension channels so a total of 80MHz, in fact your status screen shows 80MHz

When you btest you should not btest on the devices, the CPU will max out, you should btest between other devices connected to both ends.

Also the signals are much too high, which will not help, try moving them so that signals are -40 or worse

You should also have the antenna gain set correctly for the LHG which is 24, this will then set the tx power correctly, and you also should be using regulatory domain.

Regards
Nick

Use pure 802.11 not nstreme (nstreme/NV2 dont work good with ARM at this moment). ARM have much more power & easy can handle BT directly on device.

Man you can't have 95Mbps both down and up, it's too much perfect.
You're bandwidth testing passing through a 100Mbps Ethernet for sure, or you have a queue set to 100Mbps.

Use pure 802.11 not nstreme (nstreme/NV2 dont work good with ARM at this moment). ARM have much more power & easy can handle BT directly on device.

I tried 802.11 and still I didn't reach the advertised 800gbps+. I only got like 200-300mbps max 1 way udp. if tcp both I am getting about 120mbps.

Welcome to the club. ARM and NV2 is not working well. Nstreme have the same problem.
viewtopic.php?f=7&t=136002

Only thing which works is 802.11. Mikrotik does not solve this problem . To be sure, he does not even comment.

It is very sad that in end of 2018 we need to use only 802.11 on new hardware. NV2 and Nstreme are slow and unusable.

When you btest you should not btest on the devices, the CPU will max out, you should btest between other devices connected to both ends.
- I took into consideration reg the maxing out of the cpu, I checked even at tcp my utilization is about only 20% on both ends. So i assume this isn't about the CPU. unless you say so then I will need to test again but I am fairly certain it is not due to cpu of MT maxing out.

When you btest you should not btest on the devices, the CPU will max out, you should btest between other devices connected to both ends.
- I took into consideration reg the maxing out of the cpu, I checked even at tcp my utilization is about only 20% on both ends. So i assume this isn't about the CPU. unless you say so then I will need to test again but I am fairly certain it is not due to cpu of MT maxing out.

Hello,

LHG 5ac has 4 cores and btest < 6.44.x is only single core. Therefore you will max out one core at < 25% total CPU usage. Since you have around 20% CPU usage while doing btest, you are therefore probably maxing out one core with the btest, giving you an artificial low result.

With 802.11 the most you will get with a one-way test would be 33-60% (or maybe up to 75% if you are really lucky) of the theoretical maximum of 866Mbps, since overhead etc is not factored into the theoretical data rate.

Several people have already said, you are not missing anything. Your expectation of the product is too high. Either use an LHG60 to get gigabit or you will have to deal with the connection you are getting.

The fact you are gettin 800+ burst rates is impressive to say the least, especially in the consumer area of the band.

So in real world result... What would be the best case UL/DL speed? 285Mbps - 649Mbps one way? how can I achieve the higher result? basically the transmitter are already side by side and made sure that there is no interference on the channel. " Lucky " means winning the silicon lottery or is it as in interference wise?

Also, Why are people here suggesting the lhg60 from the spec sheet it is almost identical to the LHG 5ac except for the band?

So in real world result... What would be the best case UL/DL speed? 285Mbps - 649Mbps one way? how can I achieve the higher result? basically the transmitter are already side by side and made sure that there is no interference on the channel. " Lucky " means winning the silicon lottery or is it as in interference wise?

Also, Why are people here suggesting the lhg60 from the spec sheet it is almost identical to the LHG 5ac except for the band?

Lucky meaning interference-wise. There is almost never a situation where there is no interference unless you are in the middle of nowhere away from all 5GHz home routers. The more 5GHz sources, the more interference, and you can get interference also from upper partials of lower frequency transmissions. The real world result will be based entirely on the interference in the area and you cannot control that because you cannot control what others are using in unlicensed bands.

The LHG60 can do 1Gbps full duplex actual throughput (both directions 1Gbps at the same time) but only for shorter runs, like under 2KM. It is so directional and short range that interference is not an issue.

Anyway So, the 100Mbps duplex is normal I guess. Would be nice that MT would show real world test of the radio link so people like me could have informed choices.

Thanks to you and all those responded. I was not expecting to achieve the 866 but somewhere in the 400s. never figured that it really far from that. This is my first time with ptp wireless thus me scratching my head. I thought it should be better with our aruba ap 225s but wasn't.

Thanks to you and all those responded. I was not expecting to achieve the 866 but somewhere in the 400s. never figured that it really far from that. This is my first time with ptp wireless thus me scratching my head. I thought it should be better with our aruba ap 225s but wasn't.

Although I haven't worked with those specific units, it is possible that you could get 200Mbps duplex or maybe a bit more (300Mbps may be too optimistic though). Your signals are too hot as well, you should turn down the tx power on both sides. You get significantly less throughput when the tx power is too high. Your signal on both ends should ideally be around like -55 or -60. In my experience with other MikroTik gear anything hotter and you start to get worse throughput.

Anyway, I will checking other hardware as I need something at least 500mbps. MT seems to be having issues with wireless ptp speeds as of the moment and seems that there is no solution yet.. viewtopic.php?f=7&t=136002&start=250
Already bought a couple of b5 and b5c. I was hoping not to go that route before as I like MT.

Anyway, I will checking other hardware as I need something at least 500mbps. MT seems to be having issues with wireless ptp speeds as of the moment and seems that there is no solution yet.. viewtopic.php?f=7&t=136002&start=250
Already bought a couple of b5 and b5c. I was hoping not to go that route before as I like MT.

What is the distance that the radios will be from each other? It is difficult to get 500Mbps full duplex from almost any 5GHz radio, except maybe over short distances. If the distance is < 2km (ideally <1.5km) the LHG60 would easily do that.

during testing it was around 10 meters. anyway I got it to a max of 400mbps ( I was maxing out at around 36 to 38MBps for file transfer ) with the most ideal condition ( very low noise, tested from 20+ dbi to 60+ , 802.11, 1 big file transfer around 3 meters apart ) I guess the other 2 protocols are having issues. I think I just found the most transfer rate of this device.

during testing it was around 10 meters. anyway I got it to a max of 400mbps ( I was maxing out at around 36 to 38MBps for file transfer ) with the most ideal condition ( very low noise, tested from 20+ dbi to 60+ , 802.11, 1 big file transfer around 3 meters apart ) I guess the other 2 protocols are having issues. I think I just found the most transfer rate of this device.

That seems pretty reasonable to me. I don't think you are going to get much better out of that device. The Mimosa will likely provide more throughput, but it is also more expensive.

Yes, I am not expecting more from these MT That is what I was expecting even before about the half of the 866mbps minus the overhead. But MT needs to fix the other 2 protocols to match or even exceed the 802.11. But what bothers me on the responses here it seems that they are not expecting it to perform more than 100mbps as it was describe as almost " perfect ".

Yes, I am not expecting more from these MT That is what I was expecting even before about the half of the 866mbps minus the overhead. But MT needs to fix the other 2 protocols to match or even exceed the 802.11. But what bothers me on the responses here it seems that they are not expecting it to perform more than 100mbps as it was describe as almost " perfect ".

I think you may have at least in part misunderstood the comments.

Redmor said "Man you can't have 95Mbps both down and up, it's too much perfect." but didn't mean that it is an amazing speed. He just thought it was strange that both were right below 100Mbps where you would run into a limit if you were btesting over 100Mbps ethernet. He thought that was too coincidental and that you must therefore have a 100Mbps ethernet limit somewhere.

mistry7 said that you are happy when you reach 200-300Mbps, which I agree with when there is a typical amount of interference.

Steveocee said that you getting 800+ burst rates was impressive but that obviously is not what you were getting and that was due to a misreading of your results.

Last couple of questions.. How about the LHG 60 why is it per claim able to do gigabit speed ( 1Gbps full duplex actual throughput (both directions 1Gbps at the same time) with basically the same hardware as the lhg 5 ac but just different frequency? Is this gigabit speed 802.11 or the other protocols as well?

Last couple of questions.. How about the LHG 60 why is it per claim able to do gigabit speed ( 1Gbps full duplex actual throughput (both directions 1Gbps at the same time) with basically the same hardware as the lhg 5 ac but just different frequency? Is this gigabit speed 802.11 or the other protocols as well?

The larger the channel width, the higher the bandwidth that a radio can put through (assuming no interference or minimal interference). The LHG60 (based on a wireless AD chip) is able to do gigabit actual throughput full duplex because the channel width for 60GHz radios is 2160MHz, which is huge compared to wireless AC's 80MHz channels (or 160MHz for gen2 AC). The LHG60 does not use regular 802.11ad but instead uses a proprietary MikroTik protocol with no name, although it uses an 802.11ad chip. New menus were added under wireless for 60GHz config, so there is no option to select "nstreme" or "nv2" or "802.11", it is hard fixed to the proprietary MikroTik 60GHz protocol.

If the LHG 5ac allowed you to choose a channel width of 2160MHz instead of only 80MHz it would doubtless deliver the same performance as the LHG60 if there were no interference, but it obviously doesn't allow that configuration, and there are few places on the planet where you could have a channel that wide with a center of 5GHz without interference.

60GHz doesn't have the interference issues because it is extremely directional and only works over a short distance - atmospheric attenuation is very high in that band, resulting in the 2km link limits. Another 60GHz source that is too far away or pointing in the wrong direction would not interfere at all, and that is a big part of the reason why they are able to get away with such huge channel sizes with 60GHz.

Links greater than 2km could work in most cases but if it rains the link could go down completely, so I wouldn't push it higher unless you are in a region where it doesn't rain often. Even with a 2km link, heavy rain could knock it offline, so if you do experience heavy rain, 1-1.5km would be a more reasonable practical limit. Correct pointing to ensure maximum signal is really critical because a few dB's could make the difference between decreased throughput in a heavy rain vs. the link going down entirely.