SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : Neural Nets - A tool for the 90's -- Ignore unavailable to you. Want to Upgrade?


To: Carl R. who wrote (305)11/26/1998 10:00:00 AM
From: Patrick Slevin  Read Replies (1) | Respond to of 871
 
I'm not as technically oriented as you are, I'm not quite certain what a BP model is and can only guess that the PII-450 is a faster chip..

Chips are incredibly cheap, by the way. My 15 year-old used some of money from caddying and bought a "build-it-wourself" machine specifying an AMD 333 chip. They ended up building it for him and the entire thing was around $650. Then he sprang for 15 more at a local electronics shop for the keyboard and he was done.

I have not had any of the 3 versions lock up on me, I can say that. Much of my problems appear to be related to loading data. I think I solved my problem from yesterday by running a set of data provider programs to rebuild the files.

I imagine you don't have this problem because you are loading the data yourself, but since I started using Pinnacle it's been a bit hairy. Pinnacle saves data in 12 folders so the path from Profit goes to one only. I'm not quite certain what to do with this; go back 11 times plus a 12th for the Reuters data?

Oddly, BioComp Tech Support does not know if the data can be packed into one folder or not. I imagine there is something I can do with DPACKER (from Pinnacle) but the effort that goes into loading the data should not be more onerous than creating systems.

I can appreciate the problems with mis-typing. I have to run it on a notebook and the keyboard is less than friendly. Because Profit is a single station program I cannot create the systems on a normal size machine and port it over to the notebook.

I strongly believe that much of my problems stem from the fact that Profit and NGO were built by a person who is interested in optimization models, a certain type of Industrial Engineering technique applied with a Computer Science background. The programs were developed really to conquer problems like traffic flow and the like. The Profit model is more of a sales aid, as they can say that if the NGO can model financial markets then it model just about anything.

No argument there; just that it's a system developed by Computer Scientists for people who can quickly grasp the nuances. Someone who trades for a living has to be hard-pressed to get answers. Simple issues like importing data should not be a roadblock. Yet even for their Tech Support person the solution to problems like this are knotty.

So the issue I am having a problem with is that it's the old story of a programmer seeing it from the way the program is built and the user more interested in the results. It boils down to communication. They think it's clear and I think it's as clear as mud.