* [Bloat] waveform bufferbloat test is GOOOD @ 2021-02-02 3:33 Dave Taht 2021-02-02 13:55 ` Jonathan Foulkes 0 siblings, 1 reply; 7+ messages in thread From: Dave Taht @ 2021-02-02 3:33 UTC (permalink / raw) To: bloat https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 I am very happy with this test, and I must admit, if it was discussed here, already, I must have missed the thread. Not only is the test well structured and fast - and some people like whisker plots more than me, my blessed cdfs, & the FAQ is brilliant. https://www.waveform.com/tools/bufferbloat#erid5407983 -- "For a successful technology, reality must take precedence over public relations, for Mother Nature cannot be fooled" - Richard Feynman dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-02 3:33 [Bloat] waveform bufferbloat test is GOOOD Dave Taht @ 2021-02-02 13:55 ` Jonathan Foulkes 2021-02-02 16:27 ` Dave Taht 0 siblings, 1 reply; 7+ messages in thread From: Jonathan Foulkes @ 2021-02-02 13:55 UTC (permalink / raw) To: Dave Taht; +Cc: bloat [-- Attachment #1: Type: text/plain, Size: 1421 bytes --] Excellent, nice to see another comprehensive test. I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 Nice to see that A+ and zeros for bloat. for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 <https://www.dslreports.com/speedtest/67140120> Cheers, Jonathan > On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: > > https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 > > I am very happy with this test, and I must admit, if it was discussed > here, already, I must have missed the thread. Not only is the test > well structured and fast - and some people like whisker plots more > than me, my blessed cdfs, & the FAQ is brilliant. > > https://www.waveform.com/tools/bufferbloat#erid5407983 > > > -- > "For a successful technology, reality must take precedence over public > relations, for Mother Nature cannot be fooled" - Richard Feynman > > dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat [-- Attachment #2: Type: text/html, Size: 2570 bytes --] ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-02 13:55 ` Jonathan Foulkes @ 2021-02-02 16:27 ` Dave Taht 2021-02-03 9:42 ` Sebastian Moeller 0 siblings, 1 reply; 7+ messages in thread From: Dave Taht @ 2021-02-02 16:27 UTC (permalink / raw) To: Jonathan Foulkes; +Cc: bloat Regrettably my joy only lasted a few hours, as there was confirmation bias. Try it with sqm off or on. Running a complete battery of tests now. On Tue, Feb 2, 2021 at 5:55 AM Jonathan Foulkes <jf@jonathanfoulkes.com> wrote: > > Excellent, nice to see another comprehensive test. > I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) > > Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: > https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 > > Nice to see that A+ and zeros for bloat. > > for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 > > Cheers, > > Jonathan > > On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: > > https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 > > I am very happy with this test, and I must admit, if it was discussed > here, already, I must have missed the thread. Not only is the test > well structured and fast - and some people like whisker plots more > than me, my blessed cdfs, & the FAQ is brilliant. > > https://www.waveform.com/tools/bufferbloat#erid5407983 > > > -- > "For a successful technology, reality must take precedence over public > relations, for Mother Nature cannot be fooled" - Richard Feynman > > dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat > > -- "For a successful technology, reality must take precedence over public relations, for Mother Nature cannot be fooled" - Richard Feynman dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-02 16:27 ` Dave Taht @ 2021-02-03 9:42 ` Sebastian Moeller 2021-02-03 11:54 ` Sina Khanifar 0 siblings, 1 reply; 7+ messages in thread From: Sebastian Moeller @ 2021-02-03 9:42 UTC (permalink / raw) To: Dave Täht; +Cc: Jonathan Foulkes, bloat Hi Dave, I wonder what is going on with the reported latencies. I see negative latency numbers indicating that the report is presenting a difference, maybe unloaded minus the direction under tests? My tests with SQM also report the unloaded latency figure ion the top right as 0ms (while my link forces A >=12ms RTT to the first IP address after my router). I would guess that there is still some development going on and hopefully that will be fixed. Best Regards Sebastian > On Feb 2, 2021, at 17:27, Dave Taht <dave.taht@gmail.com> wrote: > > Regrettably my joy only lasted a few hours, as there was confirmation > bias. Try it with sqm off or on. Running a complete battery of tests > now. > > > > > > On Tue, Feb 2, 2021 at 5:55 AM Jonathan Foulkes <jf@jonathanfoulkes.com> wrote: >> >> Excellent, nice to see another comprehensive test. >> I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) >> >> Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: >> https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 >> >> Nice to see that A+ and zeros for bloat. >> >> for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 >> >> Cheers, >> >> Jonathan >> >> On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: >> >> https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 >> >> I am very happy with this test, and I must admit, if it was discussed >> here, already, I must have missed the thread. Not only is the test >> well structured and fast - and some people like whisker plots more >> than me, my blessed cdfs, & the FAQ is brilliant. >> >> https://www.waveform.com/tools/bufferbloat#erid5407983 >> >> >> -- >> "For a successful technology, reality must take precedence over public >> relations, for Mother Nature cannot be fooled" - Richard Feynman >> >> dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 >> _______________________________________________ >> Bloat mailing list >> Bloat@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/bloat >> >> > > > -- > "For a successful technology, reality must take precedence over public > relations, for Mother Nature cannot be fooled" - Richard Feynman > > dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-03 9:42 ` Sebastian Moeller @ 2021-02-03 11:54 ` Sina Khanifar 2021-02-03 21:03 ` Sebastian Moeller 2021-02-04 9:43 ` Michael Yartys 0 siblings, 2 replies; 7+ messages in thread From: Sina Khanifar @ 2021-02-03 11:54 UTC (permalink / raw) To: bloat Hi all, It's funny that this thread should pop up now … I drafted a long email to send to the list two nights ago. I was planning to announce that we were almost ready to actually launch our updated bufferbloat test, but then we found some pesky bugs that we're now working to squash. We should have them all sorted out this week or early next week, and I'll finally send my longer email to the group that covers all the changes we made since we shared an early beta a couple of months back. Best, Sina. On Wed, Feb 3, 2021 at 1:42 AM Sebastian Moeller <moeller0@gmx.de> wrote: > > Hi Dave, > > > I wonder what is going on with the reported latencies. I see negative latency numbers indicating that the report is presenting a difference, maybe unloaded minus the direction under tests? My tests with SQM also report the unloaded latency figure ion the top right as 0ms (while my link forces A >=12ms RTT to the first IP address after my router). I would guess that there is still some development going on and hopefully that will be fixed. > > Best Regards > Sebastian > > > > > > On Feb 2, 2021, at 17:27, Dave Taht <dave.taht@gmail.com> wrote: > > > > Regrettably my joy only lasted a few hours, as there was confirmation > > bias. Try it with sqm off or on. Running a complete battery of tests > > now. > > > > > > > > > > > > On Tue, Feb 2, 2021 at 5:55 AM Jonathan Foulkes <jf@jonathanfoulkes.com> wrote: > >> > >> Excellent, nice to see another comprehensive test. > >> I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) > >> > >> Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: > >> https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 > >> > >> Nice to see that A+ and zeros for bloat. > >> > >> for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 > >> > >> Cheers, > >> > >> Jonathan > >> > >> On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: > >> > >> https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 > >> > >> I am very happy with this test, and I must admit, if it was discussed > >> here, already, I must have missed the thread. Not only is the test > >> well structured and fast - and some people like whisker plots more > >> than me, my blessed cdfs, & the FAQ is brilliant. > >> > >> https://www.waveform.com/tools/bufferbloat#erid5407983 > >> > >> > >> -- > >> "For a successful technology, reality must take precedence over public > >> relations, for Mother Nature cannot be fooled" - Richard Feynman > >> > >> dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 > >> _______________________________________________ > >> Bloat mailing list > >> Bloat@lists.bufferbloat.net > >> https://lists.bufferbloat.net/listinfo/bloat > >> > >> > > > > > > -- > > "For a successful technology, reality must take precedence over public > > relations, for Mother Nature cannot be fooled" - Richard Feynman > > > > dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 > > _______________________________________________ > > Bloat mailing list > > Bloat@lists.bufferbloat.net > > https://lists.bufferbloat.net/listinfo/bloat > > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-03 11:54 ` Sina Khanifar @ 2021-02-03 21:03 ` Sebastian Moeller 2021-02-04 9:43 ` Michael Yartys 1 sibling, 0 replies; 7+ messages in thread From: Sebastian Moeller @ 2021-02-03 21:03 UTC (permalink / raw) To: Sina Khanifar; +Cc: bloat, Dave Täht @Dave, using Chrome instead of Safari I get this with SQM: https://www.waveform.com/tools/bufferbloat?test-id=ee977631-2f50-4105-b717-5066cbf3ddca Unloaded: 15 ms Download Active: +0 ms Upload Active: +0 m this with SQM disabled: https://www.waveform.com/tools/bufferbloat?test-id=0959dcfa-04e2-435c-9924-0d0438420597 Unloaded: 15 ms Download Active: +14 ms Upload Active: +50 m and this with SQM enabled again: https://www.waveform.com/tools/bufferbloat?test-id=154ee895-108e-4bcf-8a08-536340fb5282 Unloaded: 15 ms Download Active: +1 ms Upload Active: +0 m all measured via an USB3 realtek gigabit-ethernet dongle on a macbook with 10.14.6. To me this looks like cake really helps, but I wonder what TCP is in use by cloudflare's CDN nodes... I had expected somewhat larger effects. Best Regards Sebastian > On Feb 3, 2021, at 12:54, Sina Khanifar <sina@waveform.com> wrote: > > Hi all, > > It's funny that this thread should pop up now … I drafted a long > email to send to the list two nights ago. I was planning to announce > that we were almost ready to actually launch our updated bufferbloat > test, but then we found some pesky bugs that we're now working to > squash. > > We should have them all sorted out this week or early next week, and > I'll finally send my longer email to the group that covers all the > changes we made since we shared an early beta a couple of months back. > > Best, > > Sina. > > On Wed, Feb 3, 2021 at 1:42 AM Sebastian Moeller <moeller0@gmx.de> wrote: >> >> Hi Dave, >> >> >> I wonder what is going on with the reported latencies. I see negative latency numbers indicating that the report is presenting a difference, maybe unloaded minus the direction under tests? My tests with SQM also report the unloaded latency figure ion the top right as 0ms (while my link forces A >=12ms RTT to the first IP address after my router). I would guess that there is still some development going on and hopefully that will be fixed. >> >> Best Regards >> Sebastian >> >> >> >> >>> On Feb 2, 2021, at 17:27, Dave Taht <dave.taht@gmail.com> wrote: >>> >>> Regrettably my joy only lasted a few hours, as there was confirmation >>> bias. Try it with sqm off or on. Running a complete battery of tests >>> now. >>> >>> >>> >>> >>> >>> On Tue, Feb 2, 2021 at 5:55 AM Jonathan Foulkes <jf@jonathanfoulkes.com> wrote: >>>> >>>> Excellent, nice to see another comprehensive test. >>>> I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) >>>> >>>> Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: >>>> https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 >>>> >>>> Nice to see that A+ and zeros for bloat. >>>> >>>> for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 >>>> >>>> Cheers, >>>> >>>> Jonathan >>>> >>>> On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: >>>> >>>> https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 >>>> >>>> I am very happy with this test, and I must admit, if it was discussed >>>> here, already, I must have missed the thread. Not only is the test >>>> well structured and fast - and some people like whisker plots more >>>> than me, my blessed cdfs, & the FAQ is brilliant. >>>> >>>> https://www.waveform.com/tools/bufferbloat#erid5407983 >>>> >>>> >>>> -- >>>> "For a successful technology, reality must take precedence over public >>>> relations, for Mother Nature cannot be fooled" - Richard Feynman >>>> >>>> dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 >>>> _______________________________________________ >>>> Bloat mailing list >>>> Bloat@lists.bufferbloat.net >>>> https://lists.bufferbloat.net/listinfo/bloat >>>> >>>> >>> >>> >>> -- >>> "For a successful technology, reality must take precedence over public >>> relations, for Mother Nature cannot be fooled" - Richard Feynman >>> >>> dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 >>> _______________________________________________ >>> Bloat mailing list >>> Bloat@lists.bufferbloat.net >>> https://lists.bufferbloat.net/listinfo/bloat >> >> _______________________________________________ >> Bloat mailing list >> Bloat@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/bloat ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Bloat] waveform bufferbloat test is GOOOD 2021-02-03 11:54 ` Sina Khanifar 2021-02-03 21:03 ` Sebastian Moeller @ 2021-02-04 9:43 ` Michael Yartys 1 sibling, 0 replies; 7+ messages in thread From: Michael Yartys @ 2021-02-04 9:43 UTC (permalink / raw) To: sina, bloat [-- Attachment #1: Type: text/plain, Size: 4420 bytes --] It seems like the bufferbloat test is having some issues on Firefox mobile on Android: https://www.waveform.com/tools/bufferbloat?test-id=54563bc4-03dc-4f99-be0f-6d969a70d71c The loaded latency is obviously wrong. I don't have this issue with Google Chrome on Android: https://www.waveform.com/tools/bufferbloat?test-id=81877531-2d9c-40f6-b440-8a8338923b1e It's also a bit strange that the unloaded latency is so much higher on Android compared to the values I get on my laptop (I get an unloaded latency of 3 ms with fast.com on Android): https://www.waveform.com/tools/bufferbloat?test-id=74727162-bbe4-4fdc-95f4-a4e0f2fcfdfe Michael -------- Original Message -------- On 3 Feb 2021, 12:54, Sina Khanifar < sina@waveform.com> wrote: > Hi all, > > It's funny that this thread should pop up now … I drafted a long > email to send to the list two nights ago. I was planning to announce > that we were almost ready to actually launch our updated bufferbloat > test, but then we found some pesky bugs that we're now working to > squash. > > We should have them all sorted out this week or early next week, and > I'll finally send my longer email to the group that covers all the > changes we made since we shared an early beta a couple of months back. > > Best, > > Sina. > > On Wed, Feb 3, 2021 at 1:42 AM Sebastian Moeller <moeller0@gmx.de> wrote: >> >> Hi Dave, >> >> >> I wonder what is going on with the reported latencies. I see negative latency numbers indicating that the report is presenting a difference, maybe unloaded minus the direction under tests? My tests with SQM also report the unloaded latency figure ion the top right as 0ms (while my link forces A >=12ms RTT to the first IP address after my router). I would guess that there is still some development going on and hopefully that will be fixed. >> >> Best Regards >> Sebastian >> >> >> >> >> > On Feb 2, 2021, at 17:27, Dave Taht <dave.taht@gmail.com> wrote: >> > >> > Regrettably my joy only lasted a few hours, as there was confirmation >> > bias. Try it with sqm off or on. Running a complete battery of tests >> > now. >> > >> > >> > >> > >> > >> > On Tue, Feb 2, 2021 at 5:55 AM Jonathan Foulkes <jf@jonathanfoulkes.com> wrote: >> >> >> >> Excellent, nice to see another comprehensive test. >> >> I also like that they list the IQrouter in the set of routers that mitigate bloat ;-) >> >> >> >> Since I recently shared the DSLreports test from my line, here is the same line tested via this new site: >> >> https://www.waveform.com/tools/bufferbloat?test-id=f10cfe50-6dd2-4ae3-9b40-dedd2b5087c6 >> >> >> >> Nice to see that A+ and zeros for bloat. >> >> >> >> for reference, the dslreports test: https://www.dslreports.com/speedtest/67140120 >> >> >> >> Cheers, >> >> >> >> Jonathan >> >> >> >> On Feb 1, 2021, at 10:33 PM, Dave Taht <dave.taht@gmail.com> wrote: >> >> >> >> https://www.waveform.com/tools/bufferbloat?test-id=bea9f548-086a-4a74-bbde-b05b33f689a2 >> >> >> >> I am very happy with this test, and I must admit, if it was discussed >> >> here, already, I must have missed the thread. Not only is the test >> >> well structured and fast - and some people like whisker plots more >> >> than me, my blessed cdfs, & the FAQ is brilliant. >> >> >> >> https://www.waveform.com/tools/bufferbloat#erid5407983 >> >> >> >> >> >> -- >> >> "For a successful technology, reality must take precedence over public >> >> relations, for Mother Nature cannot be fooled" - Richard Feynman >> >> >> >> dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 >> >> _______________________________________________ >> >> Bloat mailing list >> >> Bloat@lists.bufferbloat.net >> >> https://lists.bufferbloat.net/listinfo/bloat >> >> >> >> >> > >> > >> > -- >> > "For a successful technology, reality must take precedence over public >> > relations, for Mother Nature cannot be fooled" - Richard Feynman >> > >> > dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729 >> > _______________________________________________ >> > Bloat mailing list >> > Bloat@lists.bufferbloat.net >> > https://lists.bufferbloat.net/listinfo/bloat >> >> _______________________________________________ >> Bloat mailing list >> Bloat@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/bloat > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat [-- Attachment #2: Type: text/html, Size: 7003 bytes --] ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2021-02-04 9:44 UTC | newest] Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-02-02 3:33 [Bloat] waveform bufferbloat test is GOOOD Dave Taht 2021-02-02 13:55 ` Jonathan Foulkes 2021-02-02 16:27 ` Dave Taht 2021-02-03 9:42 ` Sebastian Moeller 2021-02-03 11:54 ` Sina Khanifar 2021-02-03 21:03 ` Sebastian Moeller 2021-02-04 9:43 ` Michael Yartys
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox