Hi. A large text file imports on my Mac [dual-processor with 4 gigs of memory] in under 8 hours. My client is on a PC, and after 16 hours, he says it is only about 1/3 done. He got a new machine to handle this so he doesn't have to go over the network. Anyone have any idea why the huge discrepancy? And, more importantly, what I can do about it?

Thanks much,

Audrey Cole

How much Ram does his "New" machine have? You can still buy pc's with 512ram. Which is IMO is the bare minimum for Filemaker to run smooth.

I am curious how big is the text file?

Experiencing some slowness here as well...

300MB.txt over 3 hrs on 1.7 GHz 512.

I know not much of a machine but common neither is 300 MB much of a file.

Thanks!

For what is worth:

- import into access2003 little under 4 mintues

- import into FM (stripped down to single field single table) 1hrs and 17 min

207MB text file with 790,225 records

hmmmm...

But does the recieving fields cause each record to have autoenter values? There is way to many loose ends to the nature of the import ...to be just more than prejudice!

--sd

Hi Søren Dyhr,

Don't get me wrong I will take FM over Access any time...except in this test.There is no prejudice just facts. No auto-enter-calcs at all...that's why I used a single field.(FM-text,Acc-text)

Import of text files in FM is just slow...always was, much like ODBC. I was just hoping things have gotten better in 8.5 comapred to 5,6.

Seems like not the case. |

There is no prejudice just facts

Where is the evidence that it isn't carefully chosen or seleted facts, made into substantiate a belief ...such thing happens, perhaps not here - but it does sometimes, when someone see it fit their purpose.

--sd

lol

I wish it was the other way arround in benifit of FM...I could certainly use it. :o

Pardon me for coming in with a side question, but it seems that most "progress bar" actions (I call them this way since the progress bar is not removable) such as Import, Export, Sort, Perform find... have a strange behavior in this: the progress bar starts very fast (I mean, even compared to other DBs), and progressively slow down, as if a buffer was not initialised properly.

If you interrput the process, data was really imported/sorted...

So what I do when I am in control of the source file is split it into several smaller files, and it is indeed faster.

But if a guru could give a hint of how it really works, this would be really nice.

Another thing (so my post is at least partly an aswer is that all auto-enter options defined in the destination table, especially lookups, slow down the import process. If you can disable them and calculate them later, it can also accelarate things.

If you can disable them and calculate them later, it can also accelarate things

As you might read above, was it my thought as well! But it isn't apparently?

--sd

Oups, sorry... catching up after some holydays :o