Recommended Posts

what about a wireless to your router.. Your ping test wired to the router is 0..  But that graph shows that many times your packet loss is so high that your internet would basically be unusable.

What your wanting to show is the packet loss happening inside your network, or is it happening via your isp or internet.. If its internet have them fix it!!  If you show wireless, since its their router again have them fix it..

Is that long test graph from a wired client or a wireless client?

Here is last week of my monitoring of my internet connection ping times.. Notice it worse is 0.07% So less than 1/10 of 1 percent.. And I quite often pretty much saturate my line..  Not only up but down..  Where you would expect some drops in pings..

packetloss.thumb.jpg.9da4bab796d1b84e7c3afad98d2eeae6.jpg

With 7% packet loss your internet experience is going to be horrible.. Some of those spikes showing much higher would be just unusable..

We know your wired client to internet is showing 7% loss, so this really really points to just your internet connection is crap..  But would be good to show them hey look, my local side is fine - so its not the client.. Both wired and wireless.. FIX IT!!!

 

I can't test the wireless network on its own. The only IP address I have is for the modem at is at 192.168.0.1 I do not show a separate IP for Wireless.

Here is what I really want to know before calling support: Why is the packet loss so high between the hours of 12:00 - 20:00 and then again from 01:00 to 06:00. What is happening at those tines? Obviously we are asleep at 0100, and I don't get up before 0630-0700 so that is just background transfers, updates, etc.

Ethernet adapter Ethernet:

   Connection-specific DNS Suffix  . :
   Link-local IPv6 Address . . . . . : fe80::4ee:b793:17a4:22ad%15
   IPv4 Address. . . . . . . . . . . : 192.168.0.2
   Subnet Mask . . . . . . . . . . . : 255.255.255.0
   Default Gateway . . . . . . . . . : 192.168.0.1

Ethernet adapter Bluetooth Network Connection:

   Media State . . . . . . . . . . . : Media disconnected
   Connection-specific DNS Suffix  . :

Ethernet adapter vEthernet (WSL (Hyper-V firewall)):

   Connection-specific DNS Suffix  . :
   Link-local IPv6 Address . . . . . : fe80::668b:1b15:65ea:e2f6%33
   IPv4 Address. . . . . . . . . . . : 172.28.208.1
   Subnet Mask . . . . . . . . . . . : 255.255.240.0
   Default Gateway . . . . . . . . . :

I ran a ping to Google via my cellphone, and here is the results:

64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
29 ttl=118 time=45.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
30 ttl=118 time=111 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
31 ttl=118 time=59.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
32 ttl=118 time=61.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
33 ttl=118 time=106 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
34 ttl=118 time=48.7 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
35 ttl=118 time=102 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
36 ttl=118 time=79.5 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
37 ttl=118 time=112 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
38 ttl=118 time=42.6 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
39 ttl=118 time=72.1 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
40 ttl=118 time=104 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
41 ttl=118 time=62.0 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
42 ttl=118 time=102 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
43 ttl=118 time=51.2 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
44 ttl=118 time=110 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
45 ttl=118 time=48.9 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
46 ttl=118 time=117 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
47 ttl=118 time=71.0 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
48 ttl=118 time=109 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
49 ttl=118 time=82.4 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
50 ttl=118 time=51.7 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
51 ttl=118 time=73.1 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
52 ttl=118 time=46.9 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
53 ttl=118 time=57.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
54 ttl=118 time=45.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
55 ttl=118 time=47.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
56 ttl=118 time=56.4 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
57 ttl=118 time=86.0 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
58 ttl=118 time=100 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
59 ttl=118 time=62.1 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
60 ttl=118 time=107 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
61 ttl=118 time=50.4 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
62 ttl=118 time=59.2 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
63 ttl=118 time=104 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
64 ttl=118 time=98.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
65 ttl=118 time=60.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
66 ttl=118 time=54.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
67 ttl=118 time=124 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
68 ttl=118 time=54.6 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
69 ttl=118 time=47.9 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
70 ttl=118 time=101 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
71 ttl=118 time=54.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
72 ttl=118 time=115 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
73 ttl=118 time=58.1 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
74 ttl=118 time=151 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
75 ttl=118 time=98.1 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
76 ttl=118 time=116 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
77 ttl=118 time=67.6 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
78 ttl=118 time=105 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
79 ttl=118 time=51.6 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
80 ttl=118 time=111 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
81 ttl=118 time=52.5 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
82 ttl=118 time=104 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
83 ttl=118 time=47.5 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
84 ttl=118 time=158 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
85 ttl=118 time=48.0 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
86 ttl=118 time=129 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
87 ttl=118 time=65.7 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
88 ttl=118 time=122 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
89 ttl=118 time=58.8 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
90 ttl=118 time=49.0 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
91 ttl=118 time=116 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
92 ttl=118 time=55.3 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
93 ttl=118 time=98.9 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
94 ttl=118 time=121 ms
64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
95 ttl=118 time=51.9 ms
--- google.com ping statistics ---
295 packets transmitted, 295 received, 0% packet loss, time 294646ms
rtt min/avg/max/mdev = 36.991/73.904/158.898/28.407 ms

I tried to ping the desktop, but it timed out, and pinging the modem yielded the same results as from the desktop

Edited by jnelsoninjax
On 30/04/2024 at 09:44, jnelsoninjax said:

The only IP address I have is for the modem at is at 192.168.0.1 I do not show a separate IP for Wireless.

why would you think its a different IP? From your phone, you sure your not on cell there because I see 0 packet loss.  Or your currently not having any issues.. From your graph you put up there are times when not showing any loss maybe this ping test is one of those times?

Your routers IP is going to be the same IP for both wireless or wired..

There is a good free test for both iphone or android from hurricane electric

https://networktools.he.net/

Also on your phone or tablet you should be able to look at the connection and get the devices, ip gateway (ip address you would want to ping, ie your router) and dns.  If not this tool will show you that info.

But your previous test to google wired was 7% loss, but your wireless connection shows zero.. That doesn't make a lot of sense... Other than the issue is internet connection, and currently its not having a problem..

 

OK, I have a bit more insight into what exactly might be the issue here. Apparently when the service was first installed, the installer was unable to access a locked box on the building where it appeared the coax lines were terminating. He did not have a key, the apartment management did not have a key, but would not allow him to cut the lock off to gain access, so he somehow ran the line another route, but per my wife, she has had lots of troubles with this service and at one point a tech came out and did some extensive testing on the lines and determined there was lots of noise, so he installed filters on the line in an attempt to correct this issue. I am willing to bet that the noise issue is not corrected and that the filters are only keeping the noise down enough to create a passable score/service.

Quite possible.. your levels could be at borderline values as well.. Call them, have them come out and fix it.. Show them your testing you have done where you see 7% packet loss.. 2.4 is never going to see 200mbps, it just not.. There is no way to get that..  You would have to be running the non official 40mhz VHT and you would have to have like 3 streams min.. Doubt your router can do 3, and never seen a client that has 3, quite often they are just 1..

You might get close with 2 streams 40 VHT and 40ns guard.. That puts you at a 300 PHY, so you could maybe see like 180 if the stars were all aligned... But makes little difference because from your graph your not running 40VHT, and I doubt phones or tablets support it, and any iot devices sure and the hell not, and even routers that allow you to turn it on are suppose to auto downgrade to 20 if they notice any issues, or clients that can't connect on it, etc. etc..  A screaming good number on 2.4 would be like 70ish..

And you don't even have access to make those changes anyway.. Because you being a user would just mess it up, just ask the support guy from your ISP :) hehehe

Also you could have the best wifi possible connecting with new 802.11be (wifi 7) , out in the boonies without another wifi in 20 miles and zero interface from anything and with 7% packet loss to the internet your going to have bad experience.

On 30/04/2024 at 10:44, jnelsoninjax said:

64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
87 ttl=118 time=65.7 ms

okey

 

On 30/04/2024 at 10:44, jnelsoninjax said:

64 bytes from den08s06-in-f14.1e100.net (142.250.72.14): icmp_seq=2
86 ttl=118 time=129 ms

too much

It got to the point this morning that I could not even browse through my phone. I called support and got someone in India most likely who swore up and down that the modem is broadcasting on channel 11, despite the fact that I can see it is on 8. He was in such a hurry to get me off the phone, I am going to call back on Monday when I get a person in the states who might listen to me...

do your test you posted before that shows what wifi it can see - you were clearly on channel 6 before.

This is your ssid is it not?

 

ssid.thumb.jpg.6b470ed3128d5ba64c58a5ec39b5766e.jpg

Clearly that would make sense - its the strongest signal from your test device that I assume you were in your in your own home.

From this test you can clearly see who is on 1, 6 or 11 the channels that make sense for 2.4, and there are a few running on 6 that are using 40mhz VHT - those are the fat bell curves.. You can see how the bleed over into the the 1 and 11 channels.

From this report it makes most sense for you to be on 11, this is the lowest level signals you are seeing.. There is someone on 6 that is trying to hide their ssid (pointless) and your seeing higher signals on 1 and 6 - lowest other signals are on 11 so it make sense for you to be on 11.

On 04/05/2024 at 20:05, BudMan said:

do your test you posted before that shows what wifi it can see - you were clearly on channel 6 before.

This is your ssid is it not?

 

ssid.thumb.jpg.6b470ed3128d5ba64c58a5ec39b5766e.jpg

Clearly that would make sense - its the strongest signal from your test device that I assume you were in your in your own home.

From this test you can clearly see who is on 1, 6 or 11 the channels that make sense for 2.4, and there are a few running on 6 that are using 40mhz VHT - those are the fat bell curves.. You can see how the bleed over into the the 1 and 11 channels.

From this report it makes most sense for you to be on 11, this is the lowest level signals you are seeing.. There is someone on 6 that is trying to hide their ssid (pointless) and your seeing higher signals on 1 and 6 - lowest other signals are on 11 so it make sense for you to be on 11.

Screenshot_20240504-203325.thumb.png.89f9f6178947e38efa8809e88f3afcbd.png

I meant channel 6, not 8, I was trying to process what they told me, besides the BS.

From that graph you should be on 11, see how the other networks other than yours go down.. 1 has highest signal, then 6 and then 11.. So if I was going to pick what channel to be on for 2.4 it would be 11.

graph.thumb.jpg.f949f285f194b90fd66ea0b8c71fe025.jpg

That one guy running 40 vht on his channel 6 is being an ass and bleeding over to both 1 and 11..  At least is low level signal.

You can clearly see what channels the other networks are on.. I drew a line to show the strength of the other signals you're seeing..  The peak of the bell is the channel they are on, and you can see how signal strength drops off.. See how the one is like a fat bell and bleeds over down to channel 2 and up to channel 10. That is the one running 40mhz vht.

The highest signal on 1 is at like 58 or something, while 11 is at something around 74 maybe.. While your signal is at 29 which is screaming strong signal..

 

Yeah - how would your signal be anything but the highest.  That is the network you connect to right.  But they called it 4-8, never in my life heard it called that..

But your issue is prob not wifi interference - its that you have large amount of packet loss to the internet.

8796e3ed-7a91-46a8-bfa8-150834e98514.thumb.png.f70e9c284fae0f4c4223c5538deb535b.png

Would it do me any good to call back to support and tell them I know that they did not change my channel as requested, and demand that it be done? Or should I just say screw it and leave it alone until I can afford a new modem?

Your wifi is not actually the problem from what I have seen its that packet loss to the internet!! I would call them to come and FIX it!!!

I doubt your new modem and router is going to fix the packet loss - unless their device is just  POS..

That test were you were wired and showed 7% packet loss average - and then the other graph showing times were the packet loss was crazy high.. Your internet experience is just going to be plain horrible with 7% even, and higher it just gets worse.. 

Show them that test, have them come out and test for themselves.. I had an issue where my upload was not anywhere close to the 50 I pay for, and was seeing like 3% packet loss.  I had them come out, they validated it was an issue - it took them 3 days to get it worked out, but there is nothing you can do on your end to fix packet loss to the internet.. That has to be done by the isp..  Now if it was on your end, bad client, bad wifi, etc.. Ok there is stuff you can do.. But when you see the packet loss you had with a wired device.. It screams the isp issue, or that router is just a POS.. Unless your client doing the testing was just a POS too..  But you has zero loss pinging your routers IP.. So that just screams its the ISP problem.. Have them come out and fix it!

So here is an update to this saga. On Sunday the modem just crapped out, would not connect to the ISP. A tech came out on Monday and determined that the cable had come unscrewed from the wall, which is odd since it is behind a shelf, but when the tech reconnected it, the modem still would not get a signal, so she swapped it out for a new one and placed a signal booster? thing on the cable line going into the modem. Everything worked great for 2 days.

On Tuesday it decided to quit working again, tech came out yesterday, but I had already fixed the issue by removing the extra booster on the cable line and just placed the coax straight into the modem, and it worked. The tech who came out yesterday is the senior tech for the area and did the installation of the wiring in this complex and he complained about how he had to run the cable due to the property managers request. He said that the booster did not need to be on the modem since the signal was within the normal range. So as of now, the packet loss has slowed way down, and I did mention it to the tech and he said it likely has to do with the way the coax is ran.

On 09/05/2024 at 16:56, BudMan said:

So what kind of packet loss you seeing now with that test you ran before with that one with a graph over time? Where it was avg 7% before.

Has your speed increased on 2.4?

I'll run a report on the morning, but it looked good when I last checked. I also got the modem on channel 11 which has almost no other traffic on it.

17mps isn't all that good for even 1 steam at 2.4... PHY of 2.4 should be 72mbps on a 20mhz vht.. Even rule of thumb 50% of that for real world you should be like 35mbps, your seeing half of that.. And your ping time jumps to 2 seconds when your downloading..  See that 2005 that is what your ping time is while your downloading.. Yeah you could expect that to go up.. But that is really high, and says your saturating the link.. And stuff is buffering for long time..

See here, mine went up to 57 while downloading, and 500 when uploading..  that 400mbps down isn't saturating my link, but the 54 up would be, since my isp plan is 500/50

IMG_737045442655.thumb.jpg.d3ebe496ccb48a591b75be0994a65eb2.jpg

I am glad you got your packet loss down, and with less packet loss you will have a better overall experience then you were before.  But you pay for 200 down right.. If you client can do 2 stream on 2.4 at 20mhz vht... You should realistically expect to see like 70ish Mbps down..  Which isn't great, but then again it 2.4.. 802.11n wifi.. And it would be a lot better than 17.. 

 

Here is an even more interesting piece of info. Yesterday, the ISP called my wife and told her that they have been monitoring the connection and have been noticing that it seems to be going down around midnight for a few hours. Of course, I knew this already and it really starts around 22:00 according to Netprob, so they are sending a tech back out to do some adjustments...

here is hoping they get it sorted.. You should really have like 0% packet loss overall - less than 1% you could prob live with but not optimal by any means..  And your 2.4 with 1 stream should be 30+mbps, and with 2 streams like 70ish..  With wire or 5ghz you should be able to see your full 200 down..

On 14/05/2024 at 09:17, BudMan said:

here is hoping they get it sorted.. You should really have like 0% packet loss overall - less than 1% you could prob live with but not optimal by any means..  And your 2.4 with 1 stream should be 30+mbps, and with 2 streams like 70ish..  With wire or 5ghz you should be able to see your full 200 down..

I just hope the tech they send out understands what I am going to attempt to explain, but who knows for sure.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Apple Watch Series 11 GPS just crashed to 30% off in this fast-moving Prime Day deal by Karthik Mudaliar The Apple Watch Series 11 is available for $279, down from its $399 list price, saving buyers $120, or 30%. Amazon labels the offer as selling fast, so the current price may not remain available for long. This GPS model features a 42mm aluminum case, an Always-On Retina LTPO3 OLED display capable of reaching up to 2,000 nits, and an Ion-X glass surface with improved scratch resistance. Apple rates Series 11 for up to 24 hours of normal use or up to 38 hours in Low Power Mode, with fast charging providing up to eight hours of use from a 15-minute charge. Health and fitness tools include sleep scoring, temperature sensing, ECG support, heart-rate alerts, workout tracking, sleep apnea notifications, and hypertension notifications, where available. The watch also carries IP6X dust resistance and 50-meter water resistance. This configuration is best suited to iPhone owners who want comprehensive health tracking, notifications, contactless payments, and workout data without stepping up to a larger or cellular-equipped model. The smaller case should also appeal to buyers who prefer a lighter watch, while the S/M band fits wrists measuring 130mm to 180mm. With the current generation now significantly below its usual retail price, this is a strong time to replace an aging Apple Watch or buy a first model without compromising on Apple’s newest health and display features. Grab the discounted Apple Watch Series 11 (sold and shipped by Amazon) Good to know This Amazon deal is U.S. specific, and not available in other regions unless specified. We only use first-party seller links (at the time of article publishing); ensure that you purchase from a first-party seller link only. Check out Today's Deals on Amazon | or our recent tech deals. Become a Prime member (for Students or SNAP) via Neowin Get Prime Access - Prime for half price (for qualifying Medicaid, EBT, SNAP) Subscribe to Prime Video, Audible Plus, Music Unlimited or Kindle Unlimited via Neowin As an Amazon Associate, we earn from qualifying purchases.
    • The laptop in the bedroom is an Acer with i7-10510U CPU. Acer's website states they will not be upgrading it so I had little choice other than disable secure boot. I know next to nothing on these matters so hopefully it will be fine.
    • GitHub removes manual model selection from Copilot free and student plans by Karthik Mudaliar GitHub is removing the ability to manually select an AI model from its Copilot Free and Student plans, making its automatic routing system the default and only way to choose a model. This means users on these tiers will no longer be able to deliberately select a particular OpenAI, Anthropic, Google, or Microsoft model for a task. In its announcement, GitHub said Copilot Auto will dynamically choose what it considers the best model for each request. Free and Student accounts will retain access to models from multiple families, although the available selection will continue to depend on the restrictions attached to each plan. GitHub did not identify a fixed pool of models that Auto will always use, and its documentation warns that model availability can change over time. GitHub describes Auto as more than a random fallback system. On supported surfaces, its task-optimization technology evaluates the complexity of a request alongside real-time information about model health and availability. Straightforward prompts can be routed to faster and less expensive models, while more demanding coding tasks may be sent to higher-cost reasoning models. The company says this approach should reduce rate limiting, latency, and failed requests. Auto generally selects one model along natural prompt-caching boundaries rather than repeatedly switching models during a session, as GitHub found that mid-session changes increased costs without producing sufficient improvements in output quality. Users can still check which model generated a response. In Copilot Chat, the information appears when hovering over an answer, while Copilot CLI and the Copilot cloud agent display the selected model alongside their output. Auto is available in Copilot Chat, Copilot CLI, and the cloud agent, with the exact implementation and release status varying between supported development environments. The latest restriction follows several months of adjustments to Copilot’s individual plans. GitHub temporarily halted new Pro, Pro+, and Student subscriptions in April as it sought to manage demand and service reliability. It later introduced token-based billing and began gradually reopening individual-plan registrations on June 17. Alongside the picker change, GitHub is retiring the “Preview” label from Microsoft-developed models. It argues that the label is no longer necessary because Auto handles model routing and models are continuously updated behind the scenes.
    • Look up 'inflation' kid. Ask an AI for the numbers between both games.
  • Recent Achievements

    • One Year In
      Philsl earned a badge
      One Year In
    • Dedicated
      Scoobystu earned a badge
      Dedicated
    • First Post
      Tom Schmidt earned a badge
      First Post
    • One Month Later
      D0nn13 earned a badge
      One Month Later
    • Rookie
      +ChiefOfNeo went up a rank
      Rookie
  • Popular Contributors

    1. 1
      +primortal
      461
    2. 2
      +Edouard
      177
    3. 3
      PsYcHoKiLLa
      124
    4. 4
      Michael Scrip
      79
    5. 5
      Xenon
      76
  • Tell a friend

    Love Neowin? Tell a friend!