StatView is a statistical software that only runs on Mac OS 8.6 to 9.2.2. SheepShaver is not a viable solution for a professional use (I have datasets with several thousands of rows). StatView is the perfect stat software for me and I know it so well that I do not feel like learning a new thing such as "R" for recents macOSs. I own several classic Macs and among them the fastest one, according to MacBench, is the PowerMac 7600 upgraded to G4 1000 MHz (PM7600 with Crescendo G4). However in real use I've found that the iBook G3 14" 900 MHz is way faster: time to import a tab delimited text file with approx 70.000 rows was 1 hr for the iBook and more than 4 for the upgraded PowerMac 7600. Similar results for other upgraded Macs, such as the G3 desktop with G4 at 700 MHz, or the B&W with G4 at 450 MHz, all slower than the iBook G3. So now the question, will a G4 tower at 1.25 GHz (the very last revision of the MDD which runs OS9) or a TiBook at 1000 MHz be significantly faster than the iBook G3 at 900 MHz? Of course Altivec is not needed nor makes any difference to StatView.
Anonymous
User login
Please support the defense of Ukraine.
Direct or via Unclutter App
Active forum topics
Recent content
Navigation
No Ads.
No Trackers.
No Social Media.
All Content Locally Hosted.
Built on Free Software.
We have complied with zero government requests for information.
The 7600 has a 40 MHz bus whereas that iBook has a 100 MHz bus. I'm guessing that's the biggest factor in the speed difference you're seeing, unless the issue is actually loading the file from disk.
Without Altivec, the G3 and G4 are very close in speed. I would expect the 1.25 GHz G4 to be your best bet. 167 MHz bus in that, too.
Indeed I need processing power, so the PM G4 1.25 might be the Mac of choice but I am now facing another issue. I have datasets with an average of 70000 (seventy thousands) rows and with the G3 the thing is doable.
Now I am trying to import a CSV dataset with 300000 (three hundred thousands) rows and even reserving to StatView 800 megabytes of RAM the system says "not enough memory".
Since SheepShaver cannot go beyond 1GB and the way it manages memory is not that consistent I tried with a G5 dual processor and Mac OS Tiger with 8 GB RAM. In classic I managed to give to StatView up to 7 GB (i.e. 7 millions kb) but I got the same message of not enough memory.
StatView can in theory manage 2^32 rows, ie 4,294,967,296 rows, but I guess that I am pushing it to its limits. Will keep you posted.
Since you said StatView is a OS9 application, it can't benefit from the additional memory capacity of the G5.
Actually in OS X Tiger it's possible to increase the memory allocation of classic applications via the info palette as in Mac OS 9.
Performance wise it is now ok, but unfortunately SW still stalls when trying to use the powerful "Formula..." tool.
Interesting, I had same issue. I had to give up on Statview when my lab switched to Window platform. My stats were all for science apps. I tried them all. SAS might have been the best but I needed publication quality graphs. I tried SigmaStat (pretty decent). SPSS was not very science friendly but good for business/surveys. I used JPM which was a scalled down version of SAS. I finally tried Graph Pad's Prism and its quick and easy addins - Instat and StatMate. I swear by them now. I get a lot of data in different forms and Prism handles all the data I need. It is not perfect but not comprable to my experience with SAS. They have friendly help via email. I often run textbook data to see if my calcs with Prism coincide with the text resulta, and they do. I also do npt like R or Minitabs. Think about Prism. I believe (at least in past) it is unprotected. It usually offers free upgrades or low cost ones. For graphics iyt provided publication quality. It is very flexible and can port to almost any platform. I think they have it on Macs.