Also waiting for a new version of the op24 firmware, with the final version of OpenWRT 24.10 baked in. Not going to install 4.7.4 for testing...
So you release one GL FW build with op24 within 3 months...
Aren't you afraid to turn your routers into TP Link bot ones with ancient op21 and kernel ?
Waiting for proper op24 release as well
There was never a promise that op24 will be updated or anything else. Right now, it's more or less some experimental thing - not the firmware of truth.
"...For customers preferring the open-source driver, we will provide a synchronized Native OpenWrt version labeled 4.x.x-opxx, based on the OpenWrt main branch with kernel version 6.6.x. The MTK SDK will be used for their 4.x version. We will continue to address bugs in the open-source version and will make it the main line if it eventually outperforms the closed-source..."
Yeah, but I would still label it as some experiment instead of the new default firmware.
I'm finding I have to restart dnsmasq when any changes are made (it bombs) on v4.6.6-op24. Any indication that this is fixed in 4.7?
true, but nobody expected to be downgraded to an even older version of OpenWRT (down from 23.05.4 to OP21)....
though about this statement by GL-iNet I'm still curious what makes op24 the bad performer now, this is something i don't see any transparency about.
Also between mt76 and MTK sdk both share the same calibration data blobs, alot of issues where in fact solved in those blobs, sure mt76 interpretes what mtk sdk does but it stays a different driver also because it is more mainstreamed for the latest kernel.
For me on a self compiled upstreamed snapshot of openwrt without gl stuff it works very nicely.
I did adhear more complaints over time with the mtk sdk but due to transparency i don't think there can be a fair and transparant comparison, which helps more what the community wants.
And i mean with unfair comparison, because it is known sdks often use all kinds of hacks and boosts which OpenWrt by standard source doesn't, that would result the mtk sdk is almost always preferred, i don't think you can compare them, like how it is stated.
we can make a statistic of how many users of GL-MT6000 want the version mt76 or MTK sdk , maybe that's how we'll reach a conclusion?
This is impossible...
Here on the forum, we are just a few users who do not represent the average user.
To be honest, I won't be surprised if most users are still using the original firmware that was installed when they bought the router.
And many users who like up-to-date open-source firmware may already be using OpenWRT.
Sorry, but that's not true. With the launch of the MT-6000 it was advertised it will have the latest version of OpenWRT. That's one of the main reasons, many customer bought this router. I think, people even have screenshots of this promise on the webpage.
And when I got the MT6000 it had the latest openwrt, open source driver firmware. But it had so many problem that they reverted back to the closed source driver. That's why we have 2 firmwares for it.
Please be careful how you reply because this is obviously an opinion, which is not necessarily supported by the GL-inet support staff. For example, GL-inet support instructed me to use the OP firmware to ensure that WDS functions correctly. The current closed source driver seems unable to allow WDS to function with other GL-inet hardware (even when sold as a bundle).
Hardware: GL-MT6000
Firmware 4.7.0-op24
It's on 160MHz and it should be on DFS channel according to this information:
But it's in a Non-DFS channel (43)
And selecting 'Auto' should result in picking a channel with less interference, but it's choosing one that is already used by a neighbor
- GL-MT6000 (4.7.0-op24) is the red one.
- GL-MT3000 (4.7.0-op24) is the blue one.
- Neighbor is the coral colour.
By the way, GL-MT3000 seems to has +6dB stronger signal (double of power). They are in the same room, distant 30cm from each other.
Hi,
i tried to install 4.7.0-op24 from my 4.7.4-op21-installation.
But it's still 4.7.4 after reboot
I tried following
- update via web interface
- update via web interface without settings
- update via sysupgrade
nothing works. It's still on 4.7.4-op21
Can somebody give me a hint?
Thanks
Try installing via uboot image, and not via sysupgrade. This should work.
When the system is started, the device has scanned all channels and selects an available, clean, and less interference channel.
You can restart MT6000, make it rescan the channel, and then see which channel it chooses.
Hi, I guess it should do this when we start the 5GHz network and not when I restart the router.
But this do not explain why it's picking a NON-DFS channel.
Ideally, it should scan the network frequently to select the less busiest frequency, because some neighbors could set their router to "manual" instead of "automatic".
The function you describe, similar to network tuning, always scan the surrounding channel conditions and automatically jump to another clean channel during 5G WiFi operation.
Since of frequent searches and hops on channels, the client wireless connection may occasionally be unstable.
The firmware is not supported this feature at present.
I will submit it to the PM for evaluation if they think the feature is required.
Thank you.
@bruce A month is now over, since my last post, and an RC for 4.7.4 OP21 is out. But for OP24 there is nothing. I have more and more the feeling, the is no active development for OP24 anymore. Correct me, if I am wrong.