>>>I was just talking to our head network guy and he explained to me how we are setting this all up. We are dealing with an internet provider that covers pretty much all of North America all connected by T3. What we are going to do is connect directly to our ISP with a burstable T1 line at head office and a 56K line at the branches. With this setup we are bypassing all the hop and hub traffic of the phone company. So, if a user in a branch office is running an app on WinFrame the traffic is travelling via 56K line to ISP server, across T3, and down T1 to WinFrame and back.
>>>
>>>By the way, the thread "Is VFP Bitmapped" states the reason VFP doesn't work well with WinFrame over phone lines.
>>
>>Thanks for the info. You may still have this bottleneck problem because the 56k line in the slowest link in the chain. If you have multiple users in that branch office all fighting over that 56k line you can expected hang times with more than 4 users. That's what we found anyway. If you test this and don't feel this bottleneck please let me know ASAP.
>>
>>Thnaks
>>
>>
>>Scott k
>
>Right now we are only testing access via RAS and the internet via dialup. We hope to have the T1 and 56K line setup and start testing within a couple of months. I will let you know the results.
>
>In the mean time I have a question about your setup you described. The T1 backbone connecting the two offices you described -- is that line direct from one office to the other? If so, what's the 56K line for? The offices aren't connected directly with t1. Thier is a 56k (full duplex) line that connects the office to a telephone company frame relay switch which in turn directs the signal to a dedicated t1 that connects to the other switch which in turn sends the signal down a 56k linr to the other office whcih has the server. We also tested with RAS using 14.4 modems and found it to be pretty good. It was a little slow but not bad, 28.8 I think would be fine. Don't forget that a 28.8 modem gives you almost 28.8k of band width. The 15k rule still applies whether you are using dialup or 56k. The only difference being dialup in dedicated bandwidth if you are not shareing modem lines. A true test for you would be to slow your modem speeds down to 9600, this would be the same as having about 5 users on the 56k line give or take slightly. You should find that the bandwidth is in direct proportion with the number of users. If you can try this let me know your results.

Scott