KBSC blogs

BRMS Performance Benchmarks 2006
G5 vs Intel
American Press and Olympic Ideals
Faulty Benchmarks on Mac Intel machines
New BRMS tools coming. Now!!
Performance New Intel Macs and G5 Macs
Performance Per Watt !!??
Mac iLife 6
Mac Users
Business Rule Conference 2005

BRMS Performance Benchmarks 2006

jowens | 02 September, 2006 11:51

Benchmarks have become a big part of computer life - whether it's comparing database performance, MIPS, G5 Mac's to Intel Macs or whatever. My business is rulebased systems and I compare one rulebase against another. A few years ago we let the major vendors call them BUSINESS Rule Management Systems. Big Mistake!! Now they think that because it's a "Business" rulebase that they shouldn't have to be held to any kind of comparison that was designed for antiquated rulebase system. Well, I still think that they should have to be rated for performance in addition to making notes about all of the bells and whistles that they have tacked on to the outer edges. There's a lot more to this world for rulebased systems than just spreadsheets, decision tables and rule trees. See http://www.kbsc.com/Performance2006.xls for our latest results. We try to update it weekly so check back every now and then to see what's new. SDG, jco.

G5 vs Intel

jowens | 08 March, 2006 17:28

OK, enough is enough. I've said over and over and over that the tests that are being run on the part of Mr. Jobs are totally unfair; they always compare a dual-core Intel 2GHz to either a single core 1.xGHz G4 or a single core 2.xGHz G5 and NEVER to a dual-core G5 that has only recently been available (before Intel) nor the Dual CPU G5, which has been available for quite some time. However, in MacWorld of April 2006 they let the cat out of the bag!! Following are the times for a SINGLE core Intel at 2GHz and a SINGLE core G5 at 2.1GHz: ******* IMovie 6: Apply B&W Effect:: iMac Intel = 2:18 and iMac G5 = 3:11, Intel Better. (Hold on - it gets better!!) iMovie 6: Apply Aged Effect:: iMac Intel = 1:43 and iMac G5 = 1:12, G5 Better iMovie 6: Apply Rain Effect:: iMac Intel = 2:41 and iMac G5 = 2:43, Intel slightly better iPhoto 6: Import 100 Filse:: iMac Intel = 1:29 and iMac G5 = 1:22, G5 slightly better iPhoto 6: Export to QuickTime:: iMac Intel = 1:36 and iMac G5 = 1:31, G5 slightly better iPhoto 6: Export to Web:: iMac Intel = 2:13 and iMac G5 = 2:12, G5 slightly better iPhoto 6: Export to File:: iMac Intel = 2:59 and iMac G5 = 2:40, G5 Better iPhoto 6: Create Zip Archive:: iMac Intel = 2:33 and iMac G5 2:53, Intel Better Doom3 v1.3.1303: Beta Demo:: iMac Intel = 34.3 and iMac G5 = 17.0, G5 Much Better iTunes 6.0.2: MP3 Encode:: iMac Intel = 2:04 and iMac G5 = 1.23, G5 Much Better Art of Illusion 2.2 Render:: iMac Intel = 3:37 and iMac G5 = 6:07, Intel Much Better ******** These were the only ones for which I could do a direct comparison on the published data. But what it DOES show is that the G5 not only holds it ground against Intel, but for the most part its better or much better and Intel. Sorry, Steve. No Cigar. Next time be fair to the IBM guys and run a REAL direct comparison.

American Press and Olympic Ideals

jowens | 25 February, 2006 12:59

I was reading about Chad Hendrik in the sports section this morning. Chad is the epitome of (Spring) Texas brashness. At the same time, the free-wheeling press seems to have forgotten that his feud with "America's First Black Athlete to Win a Winter Games Medal" - Shani Davis - was because Shani Davis refused to "play along with the unwritten rules" that said that the top four men from a country should compete in the relay. Shani Davis was to compete in the 1500 meter speed skating two DAYS later and decided that he didn't want to "chance" something happening in the team event that would jeopardize his chances in his big race. ***** Sure enough, without Shani Davis, the men's team lost to the Italian team (that eventually won the Gold) and THAT is what steamed Chad. Well, folks, it steamed me as well. And, yet, there has been no press concerning Shani's overwhelming selfishness. Just about the feud (without telling why) and Chad's churlishness at losing. I hate to lose as well. As General George Patton said, "American has always loved a winner and always will. The fastest runner. The strongest man. The best ball player. I wouldn't give two hoots in Hell for someone who lost and laughed about it. Losing is despicable. Shameful." Well, George, guess what? American now touts the selfish player (Shani Davis) and disparges the team play (Chad Hendrik) because he also hates to lose. Poor Chad. Only won a single Gold, a single Silver and a single Bronze. So far. But he tried to win five. And tried with the team spirit. But you can't win five without the relay. Shani Davis? Oh, right. He went on to win his single Gold medal for that event for which he as "saving" himself and letting his teammates take it on the chin. ***** And one more thing: Why has the American press taken Shash Cohen to task for not winning the Gold in Figure Skating? GOOD GRIEF!! She was skating injured, for Pete's sake!! She was only a few hundredths of a point in front of the Russian skater (who ended up third) when the short skating routine was over. And STILL she beat her Russian contender by displaying real grace and talent; even though she fell twice at the very beginning of the program trying triple jumps that her injured leg would not let her complete. BUT, she did complete several triple jumps later in the routine. Talk ABOUT gutsy. Well, it just proves that the American Press sucks! No support for those athletes who are really gutting it out and always criticizing the wrong guy for the wrong reasons. Thank God I never went down that route and had to depend on the press and fickle fans. ******* Copyright (c) KBSC 2006

Faulty Benchmarks on Mac Intel machines

jowens | 24 February, 2006 10:43

One more time - it seems that, like our fearless leader in Silly Valley (Mr. Steve Jobs) that we are still comparing apples to oranges to peaches to plums to... Windows XP Media Edition has not been proven to handle dual CPU's properly. Windows XP running on an Intel 530 HT with Hyper-threading does seem to handle threads better than a regular Intel CPU. A more fair (fairer?) comparison of machines would be the dual-core-Intel-Mac versus dual-core-G5-Mac versus dual-core-Intel with either/or (Windows SP 64 or Windows Server 2006) versus [dual-core or dual CPU] AMD running either/or (Windows SP 64 or Windows Server 2006). Now, that would be comparing everything the same, more or less. But, really, a single 1.5 GHz G4 versus a 2.x GHz dual-core Intel? Isn't that kind of cheating to try and pump up new sales for the bastardized version of the Mac? Yes, I know that the G4 days are limited and probably G5 right after that. But that gives me about a year or so to make my point. Right now my dual-CPU G5 Mac is the fastest thing that I have for rendering movies or running AI benchmarks. Now, for a shameless AI plug among all of the Web-Weavers and J2EE-Gurim and Ant-Meisters, check out http://www.kbsc.com/Performance2000-2005.xls for benchmarks previous to all of this dual-core stuff. New benchmarks will be posted later this quarter or the first part of next quarter to http://www.kbsc.com/Performance2006.xls. I'm still trying to get something set up that will enable a more direct comparisons of BRMS (rulebased systems to old AI geeks) tools. Enjoy. SDG jco Copyright (c) KBSC 2006

New BRMS tools coming. Now!!

jowens | 18 February, 2006 20:29

I have in the past had the distinct opportunity of reviewing BRMS (Business Rule Management Systems) - rulebased systems with lots of outboard stuff - for various vendors. They even let me say bad things about them so long as it's fair. But right now there are vendors coming out with new and improved tools, especially ILOG (JRules, Rules for C/C++ and ILOG Rules for .NET) and Fair Isaac (Blaze Advisor and Blaze Advisor for .NET). Both are showing great speed improvements, more features, better reporting, easier to use GUI's for the Business Analysts, etc., etc. And now with such companies as InRules (.NET only), PegaRules, Versata, OpenRules, Jess and others coming out with "new and improved" versions, there's never been a better time to check out why some companies are suddenly making fantastic strides in software flexibility and accuracy than now. Since InfoWorld publishes most of my articles, the hard line reviews will probably appear there first. So, Every Monday check out http://www.infoworld.com/testcenter for the latest and greatest. Sometimes my stuff is there - usually not. But even if it isn't, you'll be kept up to date with some of the real-world tests on vendors. Copyright (c) KBSC 2006

Performance New Intel Macs and G5 Macs

jowens | 18 February, 2006 20:17

OK, just went through the Mac magazine that came in the mail yesterday. More speed comparisons. The old "cluky" old G5 (single core) actually kept pace with the "new and improved" dual-core Intel, both running at about 2.1 GHz. But, I ask you: Is that really a fair comparison or just Steve Jobs (and his servile minions) trying to promote an inferior product? Why NOT compare the dual-core Intel with the dual-core G5 and the Quad-Intel with the Quad G5? if the single-core G5 is keeping up with the dual-core Intel, then it seems logical that the dual-core G5 should blow the doors off of the dual-core Intel. And, one other thing: When I'm running my Dual G5 (the original stuff) i don't have to worry about a Universal or pure Mac software problem. There aren't any. So, until I get a really, really good reason to switch, I think I'd rather fight. (If you're old enough to remember that commercial, you're too old to be reading my blog. Just kidding! :-) Now, all of you who read this: Put the pressure on Apple and Intel to do FAIR comparisons, not some namby-pamby junk that compares Apples and Intels. (Intels are rough-skin Limes - suitable only for those who don't really appreciate a fine Cuba Libra.) Finally, both the software and the OS should be written to take advantage of dual CPU's, and most of the Intel stuff isn't. SDG jco Copyright (c) KBSC 2006

Performance Per Watt !!??

jowens | 16 February, 2006 17:48

Come on now! Performance per Watt? When did this become a measurement??? The article at http://www.integratedmar.com/ereportsb/story.cfm?item=230 tries to make this a selling point. As a matter of fact, here is a quote from the article: "According to Tau Leng, director of marketing for San Jose, Calif.-based Supermicro, these new Sossaman-based solutions deliver top performance per watt and optimized price-performance ratio. "This is a very important product for Intel, for us, and for our customers as it's the first (Intel) processor offering a low power solution . . . it features two sockets, each dual-core so with four ports on the motherboard and each socket is using 31 watts compared to 130 watts (previously), so that's a significant reduction of power consumption," he explained. "Power consumption is an important issue with server farms where you have hundreds of servers running and if you were to calculate the cost of running (less efficient servers), it's not hard to see the benefit." "Cooling also benefits from less power as it's equal to less heat and therefore cooling becomes easier." OK. But is that a really valid reason for buying a server??? If lowest price is one of the criteria, then how about a VIC-20 or Commodore-64? OK, they're older 8-bit machines but I'm sure that they could give great performance per watt. Or maybe not. But, who CARES!!?? If one server does the job and needs cooling and another server that does not do the job does not need cooling, guess which one I would select? Sorry, Charile, but when a mainframe or a Sun Star 15 is needed, you cannot substitute ANY of the Intel nor AMD boxes. As a matter of fact, Tom Yaeger at InfoWorld also wrote an entire column on this subject way back in August of 2005. See http://weblog.infoworld.com/yager/archives/2005/08/intel_performan.html for his article. And yet, they keep on and on and on... TTFN Copyright (c) KBSC 2006

Mac iLife 6

jowens | 16 February, 2006 14:07

Just went through (fairly in-depth) review of the Mac iLife 6. For only $99 you can get a family pack (of three) that should allow you to upgrade most of your machines that you use. The iMovie HD and the iDVD HD are really, really impressive. From what i saw today, and for most of you this is old news, you can do almost anything you want with a movie, complete with some really cool titles, credits, chapter headings, as well as add background music to otherwise dull video, add weird sound effects, add lots of cool video effects. And, if that isn't enough, for about $29 you can get some add on packages that allow you to straighten out those shakey, hand-held camera shots and several hundred other things. Almost pro stuff. Of course the semi-pro stuff costs $300 and the real pro stuff costs $1,200. Still, if you want to just play around and have fun, it's priced right. if you're a pro and need all of the pro stuff, then $1,200 is chump change for really pro software. And, the best part is that most Apple Stores have free classes on iMovie HD and iDVD HD. Which is what happend to me when i went to pick up some stuff at the Apple store and got sucked into the vortex of a free information. Copyright (c) KBSC 2006

Mac Users

jowens | 13 February, 2006 11:15

OK, enough is enough. Mac users have always been on the hind teat, but this time the Federal Government, the same ones who said that Microsoft has a monopoly on the software and the PC industry, has a quote system that is Microsoft ONLY. No Java. No Unix. No Mac. Only PC's need apply. The whole story is on the Washington Post at and it's really disheartening to see this kind of thing happen to those of us who prefer Unix (a truly universal OS) and Mac (almost universal OS). Well, once more: Enough is Enough. Write you congressman. Write you senator. Call the governor. Call the mayor. To Arms! To Arms! The Redmonites are coming!! Copyright (c) KBSC 2006

Business Rule Conference 2005

jowens | 10 February, 2006 15:49

Imagine, if you will, a technical, national conference where there are lots (25 to be exact) Vendors BUT none of the vendors actually give any of the presentations. At the recent November 2005 Business Rules Conference in Orlando, FL, they did just that. Actually, there were more than 25 vendors but only 25 had booth space. All of the presentations were given and explained in the detail by the ultimate, end customers themselves complete with Q&A from packed audiences, mostly dealing with, "HOW did you do that AND are those really the ROI results or just estimates?" Most new attendees were totally incredulous (may I say disbelieving) about the results for re-structuring and using a decent BRMS tool. Why, you might ask, is this important to me? Most of us can see the importance of attending a supply chain management conference, a Java conference, a J2EE/EJB conference or something along those lines. If your company already uses some kind of Business Rule Management System (BRMS - or rulebased system) then you might be aware of the advantages of using such a tool. Or not. If your company does not use a BRMS at this time, then you are rapidly falling behind the leading (not bleeding) edge of technology. One personal note: I have YET to initiate a BRMS for a company where they did not fully believe that they already knew their own business logic and work flow BUT when we began to implement the rules we discovered many, many errors in logic, procedures, work flow and documentation. Time after time, they business guys had to clean house BEFORE the geeks could help out with the implementation of the BRMS. One of the biggest impressions that I got out of the conference was that most newbies to the Rule Conference found it hard to believe that with most of the companies IT and Business were finally communicating and (almost) walking hand-in-hand. Some of the most unbelievable things that we heard was, "We are able to do analysis projects that used to cost $1M and do them now for only $50,000." (page 17 of Ron Ross 6/11/05) "It used to take two months to train someone how to do a certain process. Today it takes about two hours on-line. All because we put intelligence into our training program." (Peter Schoenrock, Equifax, 7/11/05) See? Intelligent rules can be put into almost any part of your company and greatly improve the measurable ROI. But, and this is one thing that repeatedly was pointed out by most of the older, more experienced, BRMS guys - it isn't easy. You can not take a procedural Java, COBOL or C++ programmers and architects and overnight turn them into declarative language programmers or architects. It takes time. It takes training. It takes experience. But, onece implemented, the ROI of such an adventure is absolutely astounding! So, why didn't the vendors have a major presence in the presentations? Because the founders of the conference felt (and rightly so, I might add) that the most important person in the process is the customer. If the customer can explain what they did and how they did it and the technology involved, then other customers might believe the incredible hype surrounding the industry. (After all, don't all salesmen lie?) But customers actually said things like, "I don't know. Let me get back to you on that one." And they did. Unbelievable. Not only that, most of those giving the presentations had in-depth knowledge of the tool that they used, their own business plans and how things had so dramatically improved. The conference itself is now in it's 9th year (they skipped 1999 because of the Y2K problems ) and they had such luminaries as Dr. Charles Forgy (the inventor of the Rete Algorithm) and John Zachman (the originator of the Zachman framework) at the conference. Dr. Forgy has been given credit with having given "commercial legs" to rulebased systems back in 1969 when he improved the response time of a rulebased system by more than 3,000 times. Zachman took the concept of using a rulebased system in the business world and came up with a framework on which to hang all of the various components of an enterprise business. Ron Ross, a widely respected author and entrepreneur, organized all of this into an annual conference where real "birds of a feather" could congregate and learn from each other. BRMS has now reached critical mass where even Microsoft, Computer Associates and Oracle are getting into the act. We could list all of the vendors, give a synopsis of the talks, and provide link upon link for pursual by our readers. However, if you go to either http://www.businessruleconference.org or to http://www.kbsc.com/brms you should be able to find most anything that you need along those lines. Suffice it to say that if you missed the conference this year, you probably should not miss it next year, even if it is going to be Washington D.C. area rather than in sunny Orlando. Copyright (c) KBSC 2006

Congratulations!

jowens | 13 April, 2005 11:59

If you can read this post, it means that the registration process was successful and that you can start blogging

September 2006
March 2006
February 2006
April 2005

General [3]
MacIntosh-Unix [4]
BRMS - Rulebased Systems [3]
World News [1]
Religion [0]