Glacial Search Speed

Mark A. Smith said:

If someone else could run these searches and time them on their computer and post the results with their computer specs it might help. Thanks.

I'll check back later, when L4 is finished indexing (again). Since the search function directly uses the index, I wouldn't expect it to work very well until the indexing is done.

BTW, is your system finished indexing? If not, maybe that's the culprit.

Mark A. Smith said:

No I have not downloaded the update to Beta 5 yet. I have a long download going on in the background, but that would affect the searches in 3.0 and 4.0.

Possibly. It depends on your setup, how much RAM, and what else you have going on the background (e.g., Google Desktop used to really slow down my old system), and how your anti-virus handles downloads (does it scan every packet, or the full file only after your downloader finishes writing it)? There are a few other memory hogs that can really bog down a system.

You could try pausing your download (if possible) and pausing or exiting some of these other programs and see what happens.

Mark A. Smith said:

If someone else could run these searches and time them on their computer and post the results with their computer specs it might help. Thanks.

It took about 20 seconds the first time searching the OT.

It took 1.8 seconds searching the entire Bible.

Repeated the search, restricting to the OT and it took 18.90 seconds.

It's just as true in L4 as it was in L3 that searching the whole Bible takes much less time than searching part of it. As I understand the reason for this is that Logos first searches the entire Bible, and then filters the search by the passage selection.

Interestingly searching the NIV took significantly longer than the NASB95, even though the latter is a rev. int.

Mark A. Smith said:

Richard DeRuiter said:

It took about 20 seconds the first time searching the OT.

It took 1.8 seconds searching the entire Bible.

Repeated the search, restricting to the OT and it took 18.90 seconds.

It's just as true in L4 as it was in L3 that searching the whole Bible takes much less time than searching part of it.

Funny thing is the search I did in the OT in 4.0 took less time than a search in the whole Bible.

I'm still looking for comparative numbers between 3.0 and 4.0. Seems we should see an increase in speed which I am not seeing. So far those who've run searches in both versions are getting faster searches in 3.0.

I'm downloading Beta 5 now and will have to re-index, then I'll check my speeds again.

As Rich says, searching part of a bible can take longer than searching the whole. But the problem is exacerbated in L4 with indexing significantly speeding library searches whilst their algorithm for partial searches is (now?) proving to be particularly inefficient. Some partial searches have taken 5-15 minutes on my laptop, with Grid searches significantly slower than Verses - see http://community.logos.com/forums/p/1832/14028.aspx#14028 (I just ran it in B4 - Verses took 8 sec and Grid took 7.5 min).

The difference in results with your phrase search is because:-

v4 has 2 extra results in Lam1:20 & 2:11 because the footnote characters upset v3's separation algorithm whilst v4 works with actual word separation.

v3 has 1 extra result at Dan 5:14 "spirit of the gods" because of v3's separation algorithm (picking up the) AND because it defaults to a stemming search whilst v4 defaults to a nostem search (you ask for "god" and you won't get "gods" unless you check Match all word forms in the dropdown under the Search icon!).

Mark A. Smith said:

3.19 seconds. Pretty fast but as I said, I want the hit counts in just the OT so wouldn't have chosen this search.

Why not?

In L3 I have often searched the whole Bible, even when I only wanted results in one book, because it was faster to do the full search and then scroll down to my desired location.

Mark A. Smith said:

I just wonder why an All Passages search should be faster. Isn't All Passages the same as Bible and Apochrypha?

Yes, kind of curious. I suppose it's because there are passages in the LXX that don't appear in the accepted "Apocrypha." Is that right? Or perhaps there's room for future expansion for books included in some canons that are excluded from both the Roman Catholic and the Protestant cannon.

Mark A. Smith said:

So, my speeds seem to be quite a bit slower than some others here, including Dave Hooton's whose computer specs are very much like mine.

This can be due to a number of factors, including whether or not you regularly defrag your hard drive, if you have Google Desktop (or MS indexing service) running in the background, how your anti-virus deals with file accessing, as well as a host of more esoteric hardware specifications, and background system services that may be running. These can be notoriously subtle to track down, let alone resolve.

But you're right. While L4 does much better at searching our full library, it still struggles in doing searches in the Bible.