PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
I can generate permutations easier than you. it's called a CARTESIAN.
SHOW US. Yes, if you have n distinct tokens, you just need to take the n-fold cartesian product of that set and weed out all nodes in which one or more of the tokens is repeated, so reducing the n^n nodes to the n! permutations. ok. 1) you have a table called N1 that has the numbers from 1 to 100 2) you have a table called N2 that has the numbers from 1 to 100 if you want to cartesian these two tables; to make every possible combination; all you need to do is SELECT N1.N AS N1, N2.N AS N2 FROM N1, N2 this will give you a nice little combination of every possible combination for N1 and N2. all you have to do is 'forget to write a join' and you get a cartesian. which.. doesn't seem all THAT cool. but pretend I have a years table and a months table. Select Y.YYYY, M.MMM FROM tblYEAR Y, tblMONTH M This would give you a nice little 'spreadsheet' of every combination of years and months. you could still filter it. you could still say 'give me all the months for years 2003, 2004, 2005 by adding a simple where clause. I'll look at the whole determinant thing this weekend; i dont have jack **** for freetime right now; it's the oldest sons' bday.. i can't believe he's nine already.. i mean WOW where does the time go? -aaron |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
i can cartesian up to a billion records without a performance problem
in the world. with olap i can cartesian as much as you want... billions of physical records with sub-second response times. without a problem. Analysis Services 2000 solved every database performance problem the world has ever known and you guys are still stuck in Excel AS with SQL 2005 made things about 10 times more powerful; and you guys aren't on that train. -Aaron |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
Harlan Grove wrote...
.... But if joins aren't necessary, I suppose 11 separate queries aren't necessary. Should be syntactically possible to use SELECT T1.N AS N1, T2.N AS N2, T3.N AS N3, T4.N AS N4, T5.N AS N5, T6.N AS N6, T7.N AS N7, T8.N AS N8, T9.N AS N9, T10.N AS N10, T11.N AS N11, T12.N AS T12 FROM T T1, T T2, T T3, T T4, T T5, T T6, T T7, T T8, T T9, T T10, T T11, T T12 WHERE ((N1<N2) AND (N1<N3) AND (N1<N4) AND (N1<N5) AND (N1<N6) AND (N1<N7) AND (N1<N8) AND (N1<N9) AND (N1<N10) AND (N1<N11) ND (N1<N12) AND (N2<N3) AND (N2<N4) AND (N2<N5) AND (N2<N6) .... Poor syntax in the WHERE clause. Me bad. I just tried this out in Access. Given T with one field named N containing long integers from 1 to 12, Access generated the permutations of 6 items using the query SELECT T1.N AS N1, T2.N AS N2, T3.N AS N3, T4.N AS N4, T5.N AS N5, T6.N AS N6 FROM T AS T1, T AS T2, T AS T3, T AS T4, T AS T5, T AS T6 WHERE ((T1.N<T2.N) AND (T1.N<T3.N) AND (T1.N<T4.N) AND (T1.N<T5.N) AND (T1.N<T6.N) AND (T2.N<T3.N) AND (T2.N<T4.N) AND (T2.N<T5.N) AND (T2.N<T6.N) AND (T3.N<T4.N) AND (T3.N<T5.N) AND (T3.N<T6.N) AND (T4.N<T5.N) AND (T4.N<T6.N) AND (T5.N<T6.N)) ORDER BY T1.N, T2.N, T3.N, T4.N, T5.N, T6.N; *BUT* Access couldn't handle a similar query to generate the permutations of 8 items using the query SELECT T1.N AS N1, T2.N AS N2, T3.N AS N3, T4.N AS N4, T5.N AS N5, T6.N AS N6, T7.N AS N7, T8.N AS N8 FROM T AS T1, T AS T2, T AS T3, T AS T4, T AS T5, T AS T6, T T7, T T8 WHERE ((T1.N<T2.N) AND (T1.N<T3.N) AND (T1.N<T4.N) AND (T1.N<T5.N) AND (T1.N<T6.N) AND (T1.N<T7.N) AND (T1.N<T8.N) AND (T2.N<T3.N) AND (T2.N<T4.N) AND (T2.N<T5.N) AND (T2.N<T6.N) AND (T2.N<T7.N) AND (T2.N<T8.N) AND (T3.N<T4.N) AND (T3.N<T5.N) AND (T3.N<T6.N) AND (T3.N<T7.N) AND (T3.N<T8.N) AND (T4.N<T5.N) AND (T4.N<T6.N) AND (T4.N<T7.N) AND (T4.N<T8.N) AND (T5.N<T6.N) AND (T5.N<T7.N) AND (T5.N<T8.N) AND (T6.N<T7.N) AND (T6.N<T8.N) AND (T7.N<T8.N)) ORDER BY T1.N, T2.N, T3.N, T4.N, T5.N, T6.N, T7.N, T8.N; It aborted with the message 'Not enough space on temporary disk.' FWIW, Excel wouldn't have an easy time with that either since it'd require 305 worksheets to hold all 20 million-odd permutations. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
SELECT T1.N AS N1, T2.N AS N2, T3.N AS N3, T4.N AS N4, T5.N AS N5, T6.N
AS N6, T7.N AS N7, T8.N AS N8, T9.N AS N9, T10.N AS N10, T11.N AS N11, T12.N AS T12 FROM T T1, T T2, T T3, T T4, T T5, T T6, T T7, T T8, T T9, T T10, T T11, T T12 WHERE N1 NOT IN (T2.N, T3.N, T4.N, T5.N, T6.N); etc that should be a little bit easier and tempspace.. I dont ever use MDB for anything in the real world; i use MSDE-- the freeware version of SQL Sever. That way; i dont have to rewrite **** ever-- i just throw it on a real server when I need to and I have a single language for DB stuff and a single language for frontend stuff. MSDE and SQL Server to say the least; these dont have the same problems as MDB. and the best part? MS Access has the best sproc design tools in the world.. i mean-- anywhere; and product.. and db product; and professional level product.. ADP against MSDE is the most rich querying environment anywhere. I mean-- it's all drag and drop. No way OLAP handles over 37 billion comparison operations with sub-second response time. The software can't magically eliminate the CPU-boundedness. YES OH YES IT DOES KIDS im sorry your stupid IT people chose cognos or any of those other piece of **** apps www.olapreport.com microsoft has by far the largest $$ of any vendor, including oracle, ibm... MS _RULES_ the olap market. and olap is the most important market in the world. that is why i dont give a flying **** about IBM and Oracle they have already become irrelevent.. I mean.. MS can do the same thing they can; bigger and better and faster and cheaper |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
ps - just for the record
http://www.sql-server-performance.com/wp_msas_9.asp if olap -- analysis services has 12 dimensions with one level each; and each level has 100 members-- it doesn't actually store ANYTHING resembling 20m records. OLAP is a generation past relational; it is multidimensional and it can eat your 2-dimensional matrices for lunch |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
it's kinda in between.. doesn't really store them; it doesn't really
generate them on the fly. kinda does both. and it's lighting fast.. i mean.. WOW well technically; you can specify which percentage of aggregates to store; and then you can store them in multiple formats; relationally or in memory.. it's kinda crazy to think about relational - olap - relational but it really really is a beautiful solution for some things. i mean.. the bottom line is that it's about 10 times more powerful than your traditional, girly-man pivotTables. it's kinda like what you do with offline cubes-- from excel generate some client side cubes-- but it's a lot more hierarchial than that.. i mean. .it gives Excel REAL drilldown. i just think that it's insanity to claim that Excel is more powerful than SQL. I mean-- TSQL is just plain faster and more powerful than VBA. Any way you look at it. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
hahaha good stuff man
yeah; i didn't have time to look into it; but im not scared of your 'super-duper complex matrices' i mean.. seriously here spreadsheets are for babies |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
pivotTables with a super-dooper backend can do better math faster than
you can and again, for the record.. all math--- anywhere in the world-- breaks down into simple operators i mean.. SQL is more powerful than excel.. it can handle real-life datasets. it can index. it can recover. you can back it up. and you can make ONE report for 100 different customers instead of 100 spreadsheets for 100 different customers. I mean-- it's all about the big picture of thigns; do you want to cut and copy and paste all day every day, the same data again and again and again? i just have seen-- firsthand-- a hundred situations where people have ugly ugly XLS infections; and the only thing to do is to throw it all away and start from scratch. with databases, you have a migration path-- you can grow a database from Access to SQL.. or from Access or Oracle; or whatever you want to do. What do you do when you have an XLS that is too complex, heavy to fit inside of Excel anymore?? YOU HAVE TO REWRITE IT FROM SCRATCH -aaron |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
wrote...
pivotTables with a super-dooper backend can do better math faster than you can OK, use a PivotTable to invert a matrix. Just show us the details. There's a remote chance databases can perform arithmetic faster than spreadsheets (unlikely though). However, if LIKE YOU some poor naive user knows squat all about how to string arithmetic operations together to, say, invert matrices, then wouldn't it be faster just to use Excel's MINVERSE rather than spend weeks ranting and not performing? Of course that assumes said poor naive user wants to do something useful UNLIKE YOU. and again, for the record.. all math--- anywhere in the world-- breaks down into simple operators True. All it takes is knowing how to string those simple operations together. Something you evidently DON'T know how to do. i mean.. SQL is more powerful than excel.. it can handle real-life datasets. it can . . . apparently do anything EXCEPT the matrix operations I've asked you to do. So is SQL deficient or is Aaron deficient or both? The world may never know about SQL, but in your case the answer is clear. What do you do when you have an XLS that is too complex, heavy to fit inside of Excel anymore?? YOU HAVE TO REWRITE IT FROM SCRATCH Not necessarily. If it's doing a lot of data management, then it's time to put the data management into a database. But it it's calculation-intensive, databases aren't necessarily the way to go. SAS, S-Plus or R, maybe, or MatLab. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
hahahahahhaha
you really are funny.. you know that? you honestly think that poeple need to use Excel with PERL or SAS you're a friggin crackhead harlan matlab.. **** kid.. yeah.. you're right.. i mean.. databases can't possibly handle the hard math; what in the hell was i thinking shhhiiiiiiiiiiiiiiiiiiitttttt harlan spreadsheets are for babies and all you excel dorks should be drinking wine out of a brown bag; living on the streets. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
|
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
you're the one that's driving the bike kid
with one flat tire; and a limit range of 65535 feet. lol i just dont agree with you at all harlan. all of your super-duper complex math is easily broken down into plus, sum-- those types of functions. and Access IS more extensible than Excel. That is the reason i live and die by access. I automate more for breakfast than you have in your whole life. -Aaron |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
wrote...
you're the one that's driving the bike kid I'll take me driving the bike over you *SITTING* in the airplane with your head up your ass clueless about what to do next so just shouting as loud as possible. i just dont agree with you at all harlan. all of your super-duper complex math is easily broken down into plus, sum-- those types of functions. .... So show us how you'd use those simple operations to invert matrices. It's simple for me because I know enough to use appropriate tools. It seems to be impossible for you because the only thing you do on the rare instances when you pull your head back out into the sunshine all you can do is rant. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
uh.. id just make a matrix and invert it
select 1/@myvalue lol i just think that it's hilariuous.. you seriously have this idea that your math is 'too complex for a database' you're not arguging that it's too complex for access.. you're not arguging that it's too complex for SQL.. you're not arguing that the math that you 'DO'-- by remembering the NAME of a function (ROFL) is too complex ****.. what do you eat for breakfast, a bull**** sandwich?? |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
wrote...
uh.. id just make a matrix and invert it select 1/@myvalue Brilliant! Aaron can invert 1x1 matrices! How about the 4x4 matrix I posed a few weeks ago? i just think that it's hilariuous.. you seriously have this idea that your math is 'too complex for a database' Too complex for (rather, ill suited to) SQL queries, yes. And evidently way too hard for you. |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
You'll appreciate this. Your worst nightmare about to come true.
http://blogs.msdn.com/excel/archive/....aspx#comments |
PLEASE READ IF YOU PROGRAM: Help Continue Visual Basic
technically; it means that excel is entering the 20th century-- as in
this puts this in about the same league as access was in 1900. access can spit **** out to a webpage in about 100 different ways. I mean-- it's all about being able to consume your data in other apps keeping it from excel to sharepoint to sharepoint to excel-- what's the frigging point kids keep your data in a database; you can CONSUME your data in 100 different places |
All times are GMT +1. The time now is 02:50 AM. |
Powered by vBulletin® Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
ExcelBanter.com