Are the reports that you blackballed Lucent/Livingston from your
magazine and ISPCON accurate?
Yes or no please. Just one word.
I ask the readers here to take ANY further lack of response as a YES and
reciprocate in kind against Boardwatch, Rickard, and ALL ORGANIZATIONS HE IS
INVOLVED IN.
Yes, this is a call for a boycott - not for the yellow journalism, which is
his right as a publisher (heh, we all have the right to publish bullshit)
but for the extreme cowardice and attempted act of censorship on the part of
someone who CLAIMS to be for free speech.
---- Karl Denninger (karl@MCS.Net)| MCSNet - Serving Chicagoland and Wisconsin http://www.mcs.net/ | T1's from $600 monthly to FULL DS-3 Service | NEW! K56Flex support on ALL modems Voice: [+1 312 803-MCS1 x219]| EXCLUSIVE NEW FEATURE ON ALL PERSONAL ACCOUNTS Fax: [+1 312 803-4929] | *SPAMBLOCK* Technology now included at no costOn Sat, Feb 21, 1998 at 01:19:39PM -0700, Jack Rickard wrote: > Gentlemen: > > I see. So the results of the Ascend test, on the Ascend web site, provide > PROOF, that our numbers, derived actually for the purposes of a call > completion rate test, are indeed totally bogus? Criminy Tim. How am I to > respond to such a statement? > > I am aware of a NUMBER of benchmarks. I have seen none that were > persuasive to me personally, in that they were done by interested parties > or they were done in very "lab" conditions. We did a very broad scale test > across not ONE ISP or in one lab using simulators, but across the 90 > largest ISPs in the United States offering dialup service. It was a mix of > V.34, x2, and K56flex ports done side by side under precisely the same > conditions and taken across a large number of calls across a month. > > I wasn't in search of any "proof" and don't have anything to "prove." We > did the test, and have announced the results. What you do with them, or > how you interpret them for your own use, is your own affair. We have no > vested interest in Ascend, Livingston, 3COM, or anyone else, could care > less which one "wins" and in fact, it would probably be to MY best interest > if all of it came out more or less equal. As best I can tell, they did > not. > > At any particular location, and under any set of different circumstances > and assumptions, ANY result is of course possible. Under the test we set > up, that's what happened. The noted anomalies of modem connect speed are > diagramed in the March issue of Boardwatch. The call completion rate study > is quite thoroughly depicted in teh Winter 98 Directory with the full > scores for all 90 dialup ISPs depicted. We aren't going to "vote" on it. > Karl Denninger's opinion of it is totally irrelevant. It doesnt' matter if > Joe Sasek thinks we're whores. We did it. We published it. And if we > didn't believe in it we would have done neither, and in fact would have no > negative reaction at all for having not done so. Quite the contrary, ALL > the minuses for Boardwatch come from publishing this data, and there were > NO minuses, and from an advertising standpoint huge pluses, to reward > sitting down, and keeping our mouths shut on this topic. I rather > consciously chose to disclose this information to our readers, fully aware > of the consequences and the inevitable repercussions. > > How it affects YOU and whether you go with Ascend, Livingston, 3COM, etc. > are not only of no concern to me, but probably none of my business. If we > know something or have such information and DON'T publish it, I feel we'd > be remiss. It has been an eye opener to see the VEHEMENT reaction to a > press release. With supposedly knowledgeable individuals commenting > caustically, nay hysterically, about SEVERE METHODOLOGY FLAWS in a test > whose methodology they have absolutely no information about whatsoever. > It's really a comment on the stakes of the battle. Everyone is a > stakeholder. ISP's that have committed to one vendor or another have a > huge investment in it. The developers of course have a huge investment in > it. We don't. The results we got indicate a HUGE variation from common > thought on this topic. And I have to publish those results. That it would > be received with some controversy was inevitable. The level of absolute > HYSTERIA I didn't quite foresee. > > I am curious as to how such variable results can be derived. And > interested. But we had already seen the results published by the people > who sell the equipment. We do know that if Livingston, Ascend, and 3COM > each publish results each showing clearly that THEIR'S is better, then > there exists some credibility problems with such tests. That you would > offer one of these as "proof" is, unfortunately, a statement as to the > state of the industry regarding these technologies. > > OUR results will be available in March. Use them, be entertained by them, > or line your birdcage with them as you see fit. > > My own sense from looking at OUR results AND their results, is that we have > an extremely variable telephone network. From some locations, to some > locations, K56flex works just fine. Likewise x2. From OUR location, to a > sample five locations from each of THE 90 largest dialup providers, x2 > offers an overwhelming advantage. Both IBM and MCI apparently came to the > same conclusion as the top five scored in our test are all in reality one > of the two, and they are 100% x2. Karl's vast experience in dialing from > hotel rooms all over the country, would seem to refute this. The world > will have to weigh between the two. > > Jack Rickard > > > > ---------- > > From: Tim Jung <tjung@igateway.net> > > To: isp-tech@isp-tech.com; isp-ceo@isp-ceo.com > > Subject: Re: [jack.rickard@boardwatch.com: Re: Benchmarks > > Date: Friday, February 20, 1998 4:18 PM > > > > Here is the benchmark URLs for the Ascend Max series terminal servers. If > > you need benchmarks. I wouldn't be so quick to say that Ascend is worse > than > > 3COM/USR Total Control racks. > > > > Ascend K56Flex pushes more data than USR Total Control X2 > > http://www.ascend.com/2456.html > > http://www.tolly.com/Pdf+AF8-Temp/27481-7284.pdf > > > > Ascend Max 4000 Benchmarks compared to TC and Shiva > > http://www.tolly.com/dynamic/Pdf+AF8-Temp/27481-7263.pdf > > > > > > Jack Rickard here is your proof that there maybe a problem with the > > +ACI-numbers+ACI- that you collected. This test shows that the average > throughput > > and total throughput on the X2 56K doesn't hold up to the Ascend K56Flex > 56K > > connections. This isn't the measure of the reported init string > connection > > which USR is known to play games with but true performance. There are > > multiple reports that the USR units just plain and simply don't scale as > > well as the USR. I'm not sure that the PM3 was out at this point, if it > was > > it hadn't been out very long. Now these are cold hard numbers which > anyone > > can look at, not some soon to be printed article or my gut feeling, but > cold > > hard numbers from an independent testing lab. When I bought my Ascend > units > > the Livingston didn't support STAC compression which is why we went with > > Ascend over Livingston, even though we already had a few Livingston > boxes. > > > > There are two things I would recommend to anyone looking to purchase one > of > > these type of units (TC, PM3, Max4048). > > 1.........Try not to have 100 different vendors for your equipment > > 2.........Try and pick one that has the features that you need and > options > > for the future. > > 3.........Try and not pick based on X2 or K56Flex anymore as it doesn't > > matter anymore. > > > > Tim Jung > > System Admin > > Internet Gateway Inc. > > tjung+AEA-igateway.net > > > > > > _____ * ISP-CEO Email Discussion List * ____ > > To Remove: mailto:remove-isp-ceo@isp-ceo.com > > To Join: mailto:join-isp-ceo@isp-ceo.com > > To Post: mailto:isp-ceo@isp-ceo.com > _____ * ISP-CEO Email Discussion List * ____ > To Remove: mailto:remove-isp-ceo@isp-ceo.com > To Join: mailto:join-isp-ceo@isp-ceo.com > To Post: mailto:isp-ceo@isp-ceo.com - To unsubscribe, email 'majordomo@livingston.com' with 'unsubscribe portmaster-users' in the body of the message. Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>