Bufferbloat on BRSK connection

Shaolin Chicken

Shaolin Chicken

Shaolin Chicken

Associate
Joined
19 Jan 2007
Posts
1,836
Location
Up a Tree
Hello everyone,
I am experiencing bufferbloat on my BRSK 150 down/150 up connection and I am hoping you can give some advice.
Should I contact BRSK about this. Will it affect my speeds if I do.
Or would you recommend I spend money on my own router instead of the BRSK supplied router.
Here are my results:- https://www.waveform.com/tools/bufferbloat?test-id=6fe27de2-f857-4ad8-913f-10eabedca0cf
There is only me that uses the connection, so does it really matter that I am getting bufferbloat?

Thank you for reading.
 
Well, good luck, but I doubt you will get very far. If you do then please advise me!
I get the same problem with ZEN Interenet.
My latency increases to over 250ms.
What I do is use Cake and limit the bandwidth of the connection slightly, to keep the latency low at all times. This amounts to keeping the upload and download bandwidth to 90% of the advertised speeds.
 
SpellowHouse said:
Well, good luck, but I doubt you will get very far. If you do then please advise me!
I get the same problem with ZEN Interenet.
My latency increases to over 250ms.
What I do is use Cake and limit the bandwidth of the connection slightly, to keep the latency low at all times. This amounts to keeping the upload and download bandwidth to 90% of the advertised speeds.
Click to expand...
Thanks SpellowHouse, I will look into Cake
edit: Ah, NO good as it doesn't work with my router
 
Last edited:
I wouldn't be surprised if you contacted support and they didn't even know what you're talking about. They will probably be totally perplexed why someone who is getting the speed they are paying for is complaining.

They won't change anything (if they did then SQM algorithms wouldn't need to exist :P ). All ISP's care about is that the user is getting the speed advertised. What level of latency that comes with is beyond their scope.

If you're the only one using the connection then it's all upto you at the end of the day if it's worth getting a router capable of fixing it.
 
Last edited:
There's some funniness with bufferbloat tests in my experience. I remember when I first got Community Fibre 1gb symmetrical, and when I had a go a the bufferbloat test, it came up with a A rating in Firefox. It was subsequently found that the tests can be heavily influenced by the browser you test it on when I tried it on Edge browser, and it gave me a C rating. So unless if you have something seriously noticeably wrong that is caused by bufferbloat, I'd pay it no mind really.
 
I've quick tested just now on 4 browsers as on same ISP

Firefox = A, Opera = C, Brave = B, Edge = B so all over the place

I've played Cod2 MP last weekend and World War Z MP Last night and had no issues at all. It might be worth having a blast on a game for an hour and see how you go. You could always turn of at the wall and Router for 5 mins and kick it back in. Unlikely to make a difference, but never know.

Just tested Chrome now and was a C. I'd try not to fixate.
 
Last edited:
Bufferbloat is a real and pressing problem. That said, in order to see its effects you need to have the pipe fully saturated and it's much more prevalent (and its affects more apparent) on slower and/or asymmetric connections. For example, 1000 down and 50 up will show effects of bloat much more than symmetric gigabit.

With 150/150 you shouldn't see too many real world issues, as it'd be fairly difficult to fully saturate the pipe in both directions. Maybe if you had a popular torrent running and some backup stuff, and then tried to game or place a video call. The solution is fairly simple (from an end-user perspective, not so much the underlying engineering) - just use CAKE or fq_codel (buy a Flint2 or similar, or run OpenWRT/OPNsense). On a symmetric fibre connection and having not experienced any issues, I'd not rush however.

As an aside, I find that the Cloudflare test and even ThinkBroadband test are more reliable than Waveform as a rule.
 
Mr Oxide said:
I've quick tested just now on 4 browsers as on same ISP

Firefox = A, Opera = C, Brave = B, Edge = B so all over the place

I've played Cod2 MP last weekend and World War Z MP Last night and had no issues at all. It might be worth having a blast on a game for an hour and see how you go. You could always turn of at the wall and Router for 5 mins and kick it back in. Unlikely to make a difference, but never know.

Just tested Chrome now and was a C. I'd try not to fixate when I tested with it.
Click to expand...
Thank you, yes I noticed Firefox gave the best result.
 
Last edited:
Rainmaker said:
Bufferbloat is a real and pressing problem. That said, in order to see its effects you need to have the pipe fully saturated and it's much more prevalent (and its affects more apparent) on slower and/or asymmetric connections. For example, 1000 down and 50 up will show effects of bloat much more than symmetric gigabit.

With 150/150 you shouldn't see too many real world issues, as it'd be fairly difficult to fully saturate the pipe in both directions. Maybe if you had a popular torrent running and some backup stuff, and then tried to game or place a video call. The solution is fairly simple (from an end-user perspective, not so much the underlying engineering) - just use CAKE or fq_codel (buy a Flint2 or similar, or run OpenWRT/OPNsense). On a symmetric fibre connection and having not experienced any issues, I'd not rush however.

As an aside, I find that the Cloudflare test and even ThinkBroadband test are more reliable than Waveform as a rule.
Click to expand...
Thank you for the detailed reply @Rainmaker. I tested on Cloudfare on good a good rating for video streaming, Average for gaming and good for video chatting.
I don't do too much online mutiplayer gaming, so will not worry too much.
If I do notice a real problem, I will buy the Flint 2 as recommended.

Changed my mind and ordered a Flint 2 :D, now I don't have to think about bufferbloat again.
 
Last edited:
Shaolin Chicken said:
Changed my mind and ordered a Flint 2 :D, now I don't have to think about bufferbloat again.
Click to expand...
Ohh as you are the only user I was just going to suggest switching your NIC port to 100Mbps to see if that made any difference as it would be a crude way to rate limit without spending any money but should still be fast enough for online games. You could then switch back to auto / 1000Mbps to get your full bandwidth when not gaming.

Anyway enjoy your new router.
 
NetMan said:
Ohh as you are the only user I was just going to suggest switching your NIC port to 100Mbps to see if that made any difference as it would be a crude way to rate limit without spending any money but should still be fast enough for online games. You could then switch back to auto / 1000Mbps to get your full bandwidth when not gaming.

Anyway enjoy your new router.
Click to expand...
Hello, thank you for the reply.
That sounds like a good idea, unfortunately it didn't work and I was still getting full speeds and therefore the same bufflerbloat rating.
 
Shaolin Chicken said:
Thank you for the detailed reply @Rainmaker. I tested on Cloudfare on good a good rating for video streaming, Average for gaming and good for video chatting.
I don't do too much online mutiplayer gaming, so will not worry too much.
If I do notice a real problem, I will buy the Flint 2 as recommended.

Changed my mind and ordered a Flint 2 :D, now I don't have to think about bufferbloat again.
Click to expand...

Cake is very good. Just set the bandwidth limits so that you get an acceptable increase in latency at maximum load and you are done. I think I set mine to 20ms max.
 
You must log in or register to reply here.
Back
Top Bottom