Loading ...
Sorry, an error occurred while loading the content.
 

[BUG] natural attacks: damage bonus wrong

Expand Messages
  • Michael Tucker
    I m reposting this with a shorter subject line, and clarifying the case since the only response so far was a nit-pick (I had sited the MM instead of the SRD)
    Message 1 of 26 , Jul 29, 2003
      I'm reposting this with a shorter subject line, and clarifying the case
      since the only response so far was a nit-pick (I had sited the MM
      instead of the SRD) (the points I sited were the same in both
      documents, by the way) and a question about equipping.

      Create a default dragon (e.g. a Young Gold Dragon).

      Go to the Inventory/Equipping tag and equip the natural weapons. They
      have default equipping locations, which were in my case:

      Bite (Natural/Primary)
      Claw (Natural/Secondary)
      Slap (Natural/Secondary)
      Wing (Natural/Secondary)

      Export the character. (I exported it to PDF using "csheet_dnd_std.fo".)

      The natural attacks all show up correctly on the output sheet, and the
      attack bonuses are all correct.

      However, the damage bonuses for the natural weapons are wrong.

      For example, a default Young Gold Dragon has STR:24, therefore a damage
      bonus of +7. Per the SRD (and MM p. 61), "Bite: ...deal the listed
      damage plus the dragon's Strength bonus...", Claw: ...deal the listed
      damage plus half the dragon's Strength bonus (round down)...", "Wing:
      ...plus half the dragon's Strength bonus...", "Tail Slap: ...plus 1 1/2
      times the dragon's Strength bonus...".

      The damage and bonuses should therefore have been:
      Bite: 2d6+7
      Claw: 1d8+3
      Slap: 1d8+10
      Wing: 1d6+3

      Yet, on the output sheet all the damage bonuses are the same (equal to
      the dragon's Strength bonus):
      Bite: 2d6+7
      Claw: 1d8+7
      Slap: 1d8+7
      Wing: 1d6+7

      This seems to be a bug.

      I'm running PCGen 5.1.6, with only PHB, DMG and MM loaded.

      (Standard disclaimer: if this has already been addressed, you have my
      apologies. I didn't see it discussed previously on this board.)

      Thanks,
      Java Kensai
    • Brass Tilde
      ... Well, you may not like this response either, but have you tried this with the latest beta version? If so, were the results the same? If not, please do
      Message 2 of 26 , Jul 29, 2003
        > I'm reposting this with a shorter subject line, and clarifying the case
        > since the only response so far was a nit-pick (I had sited the MM
        > instead of the SRD) (the points I sited were the same in both
        > documents, by the way) and a question about equipping.
        >
        > Create a default dragon (e.g. a Young Gold Dragon).

        > However, the damage bonuses for the natural weapons are wrong.

        > I'm running PCGen 5.1.6, with only PHB, DMG and MM loaded.

        Well, you may not like this response either, but have you tried this with
        the latest beta version? If so, were the results the same? If not, please
        do so, as there have been 4 beta releases and 1 production release since
        5.16. This problem may have been fixed in the interim.
      • Michael Tucker
        ... *laugh* You re right, I don t like that response, although I admit that you have a valid point. :-) No, I haven t tried this with the latest beta version.
        Message 3 of 26 , Jul 29, 2003
          On Tuesday, July 29, 2003, at 12:29 PM, Brass Tilde wrote:

          >> I'm reposting this with a shorter subject line, and clarifying the
          >> case
          >> since the only response so far was a nit-pick (I had sited the MM
          >> instead of the SRD) (the points I sited were the same in both
          >> documents, by the way) and a question about equipping.
          >>
          >> Create a default dragon (e.g. a Young Gold Dragon).
          >
          >> However, the damage bonuses for the natural weapons are wrong.
          >
          >> I'm running PCGen 5.1.6, with only PHB, DMG and MM loaded.
          >
          > Well, you may not like this response either, but have you tried this
          > with
          > the latest beta version? If so, were the results the same? If not,
          > please
          > do so, as there have been 4 beta releases and 1 production release
          > since
          > 5.16. This problem may have been fixed in the interim.
          >

          *laugh* You're right, I don't like that response, although I admit that
          you have a valid point. :-)

          No, I haven't tried this with the latest beta version. I tried it with
          the version I sited.

          [Sidebar that has nothing to do with the topic of this message: There
          have been a number of bugs introduced since 5.1.6, most notably the
          Mass Slow spell that was cast on PCGen somewhere around 5.1.7-5.1.8
          (which is why I'm still at 5.1.6). Although the development team has
          been busy adding many wonderful features since then, and there's been
          much effort in the realm of overall speed improvement, I haven't seen
          anyone say they were going to try to figure out what specific FUBAR
          occurred somewhere around 5.1.7-5.1.8, and fix *that* bug. (I would
          think that a code comparison would bring the offending code to light,
          but what do I know.) The bonus system has been completely redone, and
          is still in a state of flux. The development team themselves have
          discussed that they aren't sure whether PCGen is in a fit state for
          beta release, considering all the work that's been going on in the past
          few weeks with most of the primary staff's attention diverted to a)
          preparing for GenCon and b) dealing with the 3.5 RSRD. For all of those
          reasons, I'm still happy with my copy of 5.1.6, until another "stable
          and healthy" release comes out.]

          I would appreciate it if someone who *is* using the latest and greatest
          version would perform my simple experiment. No offense, but it probably
          wouldn't take much more time for someone to do so, than it would take
          to write an unhelpful response to my request for information. (Although
          it would take me a lot more time to download the latest version over my
          dialup connection, install it, etc., before I could even approach
          trying this out.)

          Java Kensai
        • Brad Franks
          I can confirm this problem *does* occur in 5.2.0 Brad ... From: Michael Tucker
          Message 4 of 26 , Jul 29, 2003
            I can confirm this problem *does* occur in 5.2.0

            Brad

            ----- Original Message -----
            From: "Michael Tucker" <mtucker@...>

            > I would appreciate it if someone who *is* using the latest and greatest
            > version would perform my simple experiment. No offense, but it probably
            > wouldn't take much more time for someone to do so, than it would take
            > to write an unhelpful response to my request for information. (Although
            > it would take me a lot more time to download the latest version over my
            > dialup connection, install it, etc., before I could even approach
            > trying this out.)
          • Felipe F. Diniz
            ... listed ... 1/2 ... ... The Dragon template does not adds/subtracts anything from the damage bonus, and looks like it should, from the SRD text. I
            Message 5 of 26 , Jul 29, 2003
              --- In pcgen@yahoogroups.com, Michael Tucker <mtucker@a...> wrote:
              >Per the SRD (and MM p. 61), "Bite: ...deal the listed
              > damage plus the dragon's Strength bonus...", Claw: ...deal the
              listed
              > damage plus half the dragon's Strength bonus (round
              down)...", "Wing:
              > ...plus half the dragon's Strength bonus...", "Tail Slap: ...plus 1
              1/2
              > times the dragon's Strength bonus...".
              <snip>
              >
              > Thanks,
              > Java Kensai


              The Dragon template does not adds/subtracts anything from the damage
              bonus, and looks like it should, from the SRD text.

              I think we should add to the Dragon Template these:

              BONUS:WEAPONPROF=Claw|DAMAGE|-STR/2
              BONUS:WEAPONPROF=Wing|DAMAGE|-STR/2
              BONUS:WEAPONPROF=Tail Slap|DAMAGE|STR/2

              When it says add half the strength bonus, we actually need to
              subtract it, because we add a full strength bonus by default. So,
              maybe the above formulae need to be adjusted a little bit because of
              the rounding down. (-STR+STR/2) should be the right way to do it, but
              i'm not sure about this.

              Can some nice TM create a PCGenDM BUG for this, so Tir and his LST
              monkey family can check it out?

              Of course, as always, Tir may have a nicer solution for this.
              Whenever he manages to crawl through his pile of e-mails, he can look
              into it.


              Felipe
              - OSCM Gibbon
            • merton_monk
              ... Just responding to the general slowdown reports we ve had in the recent past. I think we addressed most of the issues in 5.3.2. I ve asked if anyone has
              Message 6 of 26 , Jul 29, 2003
                --- In pcgen@yahoogroups.com, Michael Tucker <mtucker@a...> wrote:
                >

                Just responding to the general 'slowdown' reports we've had in the
                recent past. I think we addressed most of the issues in 5.3.2. I've
                asked if anyone has any specific slow-downs that they hadn't noticed
                earlier. I never got any response, so at this point I'm assuming
                that most people are happy with PCGen's current speed and will be
                ecstatic when they see just how the bonus code revamp will improve
                performance. If anyone thinks that 5.3.2 is slow, and can give a
                detailed description of what exactly is slower than some previous
                version (like 5.1.7), I'd appreciate it. On my machine PCGen runs
                very quickly, though mine might have a bit more muscle than the
                average user. :)


                -Bryan
              • Jayme Cox
                Per the SRD (and MM p. 61), Bite: ...deal the listed damage plus the dragon s Strength bonus... , Claw: ...deal the listed damage plus half the dragon s
                Message 7 of 26 , Jul 29, 2003
                  Per the SRD (and MM p. 61),
                  Bite: ...deal the listed damage plus the dragon's Strength bonus...",
                  Claw: ...deal the listed damage plus half the dragon's Strength bonus
                  Wing: ...plus half the dragon's Strength bonus..."
                  Tail Slap: ...plus 1 and 1/2 times the dragon's Strength bonus...".


                  I've checked in code to CVS which fixes all the above attacks except "Tail Slap"


                  Fixing the "Tail Slap" will require a .lst file change:

                  > BONUS:WEAPONPROF=Tail Slap|DAMAGE|STR

                  Since all secondary weapons only get 1/2 STR bonus to damage, the Tail Slap
                  currently has 1/2 STR bonus, so it needs STR bonus added.

                  --Jayme

                  PS: Please post the tracker number here, so we know it's been entered, Thanks!
                • Felipe F. Diniz
                  ... Is this true for all creatures? At least, is it the default behavior? Didn t know that. But anyway, thanks for being so fast in this one! ... Felipe
                  Message 8 of 26 , Jul 29, 2003
                    --- In pcgen@yahoogroups.com, "Jayme Cox" <jayme@r...> wrote:
                    > Since all secondary weapons only get 1/2 STR bonus to damage, ...


                    Is this true for all creatures? At least, is it the default behavior?
                    Didn't know that. But anyway, thanks for being so fast in this one!



                    >
                    > --Jayme


                    Felipe
                  • Michael Tucker
                    ... Cool! I hadn t realized before that (most) secondary attacks deal 1/2 STR bonus to damage. I flipped through my MM looking for a counterexample, and didn t
                    Message 9 of 26 , Jul 29, 2003
                      On Tuesday, July 29, 2003, at 03:17 PM, Jayme Cox wrote:
                      > Per the SRD (and MM p. 61),
                      > Bite: ...deal the listed damage plus the dragon's Strength bonus...",
                      > Claw: ...deal the listed damage plus half the dragon's Strength bonus
                      > Wing: ...plus half the dragon's Strength bonus..."
                      > Tail Slap: ...plus 1 and 1/2 times the dragon's Strength bonus...".
                      >
                      >
                      > I've checked in code to CVS which fixes all the above attacks except
                      > "Tail Slap"
                      >
                      >
                      > Fixing the "Tail Slap" will require a .lst file change:
                      >
                      >> BONUS:WEAPONPROF=Tail Slap|DAMAGE|STR
                      >
                      > Since all secondary weapons only get 1/2 STR bonus to damage, the Tail
                      > Slap
                      > currently has 1/2 STR bonus, so it needs STR bonus added.
                      >
                      > --Jayme
                      >
                      > PS: Please post the tracker number here, so we know it's been entered,
                      > Thanks!
                      >

                      Cool! I hadn't realized before that (most) secondary attacks deal 1/2
                      STR bonus to damage. I flipped through my MM looking for a
                      counterexample, and didn't find one (other than these dragon attacks).
                      That's neat to know!

                      By the way: whoever fixes this, please be sure to account for Tail
                      Slap, Crush *and* Tail Sweep. These *all* deal "listed damage plus 1
                      1/2 times the dragon's Strength bonus". Also, note that in PCGen LST
                      file parlance, these weapons are referred to as "Slap", "Crush" and
                      "Sweep" (respectively), so the WEAPONPROF example above probably won't
                      work (since it specifies "Tail Slap").

                      Thanks, Jayme!

                      Java Kensai
                    • Jayme Cox
                      ... This bug was caused by a very specific series of code changes and was exacerbated by differences between Java 1.3 and 1.4. The fix was a massive rewrite of
                      Message 10 of 26 , Jul 29, 2003
                        > have been a number of bugs introduced since 5.1.6, most notably the
                        > Mass Slow spell that was cast on PCGen somewhere around 5.1.7-5.1.8
                        > (which is why I'm still at 5.1.6).

                        This bug was caused by a very specific series of code changes and was
                        exacerbated by differences between Java 1.3 and 1.4. The fix was a massive
                        rewrite of all the bonus handling code. It took this long because we are all
                        volunteers, with only so much time in a day to work and play....

                        > The development team themselves have discussed that they aren't
                        > sure whether PCGen is in a fit state for beta release, considering
                        > all the work that's been going on in the past few weeks with most

                        Yes, but without the testing of our users, it will never get to that stable
                        point. I only run a limited number of tests after a change, exercising the
                        portion of code I changed. Without the user base throwing odd things at the
                        code, it takes that much longer to find and fix bugs.

                        If you have the patience, please download the latest build from:
                        http://rpg.plambert.net/pcgen
                        (you only need the pcgen.jar file) and the new pdf stuff:
                        http://prdownloads.sourceforge.net/pcgen/pdf_new.zip?download

                        A total of 7 MB isn't to harsh, even over a dialup...
                      • David
                        If I have my GMT s and EST s right, the builds started failing (pcgen.jar) just after this note, -- david
                        Message 11 of 26 , Jul 29, 2003
                          If I have my GMT's and EST's right, the builds started failing
                          (pcgen.jar) just after this note,

                          -- david

                          > -----Original Message-----
                          > From: Jayme Cox [mailto:jayme@...]
                          > Sent: Tuesday, 29 July, 2003 5:37 PM
                          > To: pcgen@yahoogroups.com
                          > Subject: [pcgen] Latest beta version
                          > If you have the patience, please download the latest build
                          > from: http://rpg.plambert.net/pcgen
                          > (you only need the pcgen.jar file) and the new pdf stuff:
                          > http://prdownloads.sourceforge.net/pcgen/pdf_n> ew.zip?download
                        • Michael Tucker
                          ... I wasn t meaning to belittle anyone, Jayme. No need to apologize. I, for one, appreciate the work that everyone does to make PCGen as great as it is.
                          Message 12 of 26 , Jul 29, 2003
                            On Tuesday, July 29, 2003, at 04:37 PM, Jayme Cox wrote:

                            >> have been a number of bugs introduced since 5.1.6, most notably the
                            >> Mass Slow spell that was cast on PCGen somewhere around 5.1.7-5.1.8
                            >> (which is why I'm still at 5.1.6).
                            >
                            > This bug was caused by a very specific series of code changes and was
                            > exacerbated by differences between Java 1.3 and 1.4. The fix was a
                            > massive
                            > rewrite of all the bonus handling code. It took this long because we
                            > are all
                            > volunteers, with only so much time in a day to work and play....
                            >

                            I wasn't meaning to belittle anyone, Jayme. No need to apologize. I,
                            for one, appreciate the work that everyone does to make PCGen as great
                            as it is. However, speaking as a professional developer, I also
                            recognize that when a project is going through "massive rewrites",
                            sometimes it's best to just be patient and use an older version (as
                            I've done) until things shake out.

                            I also understand about all-volunteer organizations. I've been very
                            active in another one for over 20 years, and have been recognized
                            numerous times for service to them (including their highest award for
                            service). *Believe me*, I know what it's like to be a volunteer, what
                            it's like to organize them, and what it's like to quietly work your
                            tail off without any thanks or recognition, other than the feeling that
                            you're doing something fun and worthwhile. I also know what it's like
                            to give so much that you burn yourself out, and just can't give any
                            more, not even to other projects. (Which is why I'm not presently
                            working on the PCGen project, other than trying to help out here on
                            these lists.)

                            [While I'm in the neighborhood of that point: Hey! All you PCGen
                            volunteers! Thanks!]

                            >> The development team themselves have discussed that they aren't
                            >> sure whether PCGen is in a fit state for beta release, considering
                            >> all the work that's been going on in the past few weeks with most
                            >
                            > Yes, but without the testing of our users, it will never get to that
                            > stable
                            > point. I only run a limited number of tests after a change, exercising
                            > the
                            > portion of code I changed. Without the user base throwing odd things
                            > at the
                            > code, it takes that much longer to find and fix bugs.
                            >

                            All very true. However, there's a difference between beta testing and a
                            "production environment". It seems a little silly to use the term
                            "production environment" in reference to a tool for administrating
                            role-playing games, but still...

                            In other words, there will always be some who "boldly go where none
                            have gone before", and eagerly download the very latest beta every time
                            there's an update, just to help shake things out. (And all of us should
                            thank them for doing so!) But, don't be surprised if the great majority
                            of folks are running slightly older versions, especially when the
                            developer's list includes threads debating whether the code is even
                            stable enough to release for *beta*.

                            > If you have the patience, please download the latest build from:
                            > http://rpg.plambert.net/pcgen
                            > (you only need the pcgen.jar file) and the new pdf stuff:
                            > http://prdownloads.sourceforge.net/pcgen/pdf_new.zip?download
                            >
                            > A total of 7 MB isn't to harsh, even over a dialup...
                            >

                            Depends on one's definition of "harsh", eh? ;-)

                            On the other hand, "what's time to a derned hog?" (If you don't get the
                            reference, it's the punch line to an old, dumb joke.) You and Bryan
                            have talked me into it. I'll download 5.3.2 and see how it runs. (At
                            about 3 k/sec throughput, that's about a 40 minute download.)

                            I will, however, go ahead and download everything. No offense, but I
                            don't believe you when you say that I'll only need the pcgen.jar file.
                            With all these updates, I take it for granted that there are some
                            interdependencies between code and data.

                            As usual, if I notice anything broken I'll try to describe it here in
                            sufficient detail that one of the monkeys can try to get their paws
                            around it.

                            Thanks again for your well-spoken thoughts, Jayme, and your work on
                            PCGen.

                            Java Kensai
                          • pjak
                            ... You probably do, but it was my fault the builds stopped working. I ve fixed it now, however, so the plambert builds should work from now on. /Jonas
                            Message 13 of 26 , Jul 29, 2003
                              --- In pcgen@yahoogroups.com, "David" <Papa-DRB@h...> wrote:
                              > If I have my GMT's and EST's right, the builds started failing
                              > (pcgen.jar) just after this note,

                              You probably do, but it was my fault the builds stopped working.

                              I've fixed it now, however, so the plambert builds should work from
                              now on.

                              /Jonas
                            • Tir Gwaith
                              Michael, without testing, I can tell you that it won t. (I ve gone over it in detail before. Note the dent at the top of my forehead from banging against a
                              Message 14 of 26 , Jul 29, 2003
                                Michael, without testing, I can tell you that it won't. (I've gone over it
                                in detail before. Note the dent at the top of my forehead from banging
                                against a wall)

                                NATURALATTACKS is set up for a default of

                                *1.5 STR if only one weapon
                                *1 STR for primary weapon if more than one
                                *.5 STR for all secondary weapons

                                It needs expansion... Waht it really needs is a conversion into actuall
                                objects of their own instead of virutal objects with extremely limited
                                capabilities, and a very nasty long tag.

                                Tir Gwaith
                                PCGen Data SB and BoD
                              • Steinar Hauan
                                ... there appears still to be a speed issue : ran the following test (cleanly rebooted machine, no other procs) (1) start pcgen (2) load mid-level/complicated
                                Message 15 of 26 , Jul 29, 2003
                                  On Tue, 29 Jul 2003, merton_monk wrote:
                                  > Just responding to the general 'slowdown' reports we've had in the
                                  > recent past. I think we addressed most of the issues in 5.3.2. I've
                                  > asked if anyone has any specific slow-downs that they hadn't noticed
                                  > earlier. I never got any response, so at this point I'm assuming
                                  > that most people are happy with PCGen's current speed

                                  there appears still to be a "speed issue":

                                  ran the following test (cleanly rebooted machine, no other procs)

                                  (1) start pcgen
                                  (2) load mid-level/complicated character
                                  (3) click on "skills" button
                                  (4) exit pcgen (without saving)

                                  measured steps 1-4 with "/usr/bin/time pcgen"

                                  v5.1.6 = total 30 secs runtime
                                  v5.3.2 = total 5 minutes runtime (=300 secs = 10x slower)
                                  approx 3 mins were spent opening the character
                                  and about 2 more after step (3) above

                                  this is beyond anything that my button-clicking would do.

                                  software: java SDK v1.4.2 for Linux on RedHat Rawhide custom.
                                  hardware: p4 2.5xghz w/1gb memory and fast (raid) disk system.

                                  more info available on request.

                                  regards,
                                  --
                                  Steinar Hauan, dept of ChemE -- hauan@...
                                  Carnegie Mellon University, Pittsburgh PA, USA
                                • Thomas Jannes
                                  Anyone aware that the autobuilds are failing again since this mail. There was one succesful build, now 10 unsuccesful in a row. ... Thomas/Chipoulou OS Mad
                                  Message 16 of 26 , Jul 30, 2003
                                    Anyone aware that the autobuilds are failing again since this mail.
                                    There was one succesful build, now 10 unsuccesful in a row.


                                    > --- In pcgen@yahoogroups.com, "David" <Papa-DRB@h...> wrote:
                                    >> If I have my GMT's and EST's right, the builds started failing
                                    >> (pcgen.jar) just after this note,

                                    > You probably do, but it was my fault the builds stopped working.

                                    > I've fixed it now, however, so the plambert builds should work from
                                    > now on.

                                    > /Jonas


                                    > Yahoo! Groups SponsorADVERTISEMENT

                                    > PCGen's release site: http://pcgen.sourceforge.net
                                    > PCGen's FAQ: http://pedertest.officeline.no/PCGen
                                    > PCGen's alpha build: http://rpg.plambert.net/pcgen
                                    > To unsubscribe from this group, send an email to:
                                    > pcgen-unsubscribe@egroups.com



                                    > Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                                    Thomas/Chipoulou
                                    OS Mad Test Monkey
                                    RSRD Wannabee
                                  • pjak
                                    ... Well, this time it wasn t my fault (phew!) but I just fixed it, so autobuilds should start working with the next compile (in a few minutes I think?) /Jonas
                                    Message 17 of 26 , Jul 30, 2003
                                      --- In pcgen@yahoogroups.com, Thomas Jannes <tjannes@v...> wrote:
                                      > Anyone aware that the autobuilds are failing again since this mail.
                                      > There was one succesful build, now 10 unsuccesful in a row.

                                      Well, this time it wasn't my fault (phew!) but I just fixed it, so
                                      autobuilds should start working with the next compile (in a few
                                      minutes I think?)

                                      /Jonas
                                    • andargor
                                      Well, I was seeing a slowdown as well, but I was chalking it up to the fact that I was testing RSRD LST files, and maybe they weren t optimized . I guess what
                                      Message 18 of 26 , Jul 30, 2003
                                        Well, I was seeing a slowdown as well, but I was chalking it up to
                                        the fact that I was testing RSRD LST files, and maybe they
                                        weren't "optimized".

                                        I guess what would help is if you would post the .pcg and we can make
                                        comparative measures. I can test 5.1.3 vs 5.3.2 readily.

                                        Andargor


                                        --- In pcgen@yahoogroups.com, Steinar Hauan <steinhau@a...> wrote:
                                        > On Tue, 29 Jul 2003, merton_monk wrote:
                                        > > Just responding to the general 'slowdown' reports we've had in the
                                        > > recent past. I think we addressed most of the issues in 5.3.2.
                                        I've
                                        > > asked if anyone has any specific slow-downs that they hadn't
                                        noticed
                                        > > earlier. I never got any response, so at this point I'm assuming
                                        > > that most people are happy with PCGen's current speed
                                        >
                                        > there appears still to be a "speed issue":
                                        >
                                        > ran the following test (cleanly rebooted machine, no other procs)
                                        >
                                        > (1) start pcgen
                                        > (2) load mid-level/complicated character
                                        > (3) click on "skills" button
                                        > (4) exit pcgen (without saving)
                                        >
                                        > measured steps 1-4 with "/usr/bin/time pcgen"
                                        >
                                        > v5.1.6 = total 30 secs runtime
                                        > v5.3.2 = total 5 minutes runtime (=300 secs = 10x slower)
                                        > approx 3 mins were spent opening the character
                                        > and about 2 more after step (3) above
                                        >
                                        > this is beyond anything that my button-clicking would do.
                                        >
                                        > software: java SDK v1.4.2 for Linux on RedHat Rawhide custom.
                                        > hardware: p4 2.5xghz w/1gb memory and fast (raid) disk system.
                                        >
                                        > more info available on request.
                                        >
                                        > regards,
                                        > --
                                        > Steinar Hauan, dept of ChemE -- hauan@c...
                                        > Carnegie Mellon University, Pittsburgh PA, USA
                                      • Vance
                                        I saw a post on this slowdown a while back and the temporoary fix was to go to: Settings: House Rules: .... and turn on .... Bypass Feat Prerequisites.
                                        Message 19 of 26 , Jul 30, 2003
                                          I saw a post on this "slowdown" a while back and the temporoary "fix" was to
                                          go to:
                                          Settings:
                                          House Rules:
                                          .... and turn on .... Bypass Feat Prerequisites.
                                          This still seems to apply to speed things up, but unfortunately defeats the
                                          purpose of the "Qualify" setting on the feats Tab.
                                          (not to mention testing new FEAT.lst's.)

                                          Possibly the slowdown can be traced to this section of code.

                                          VR


                                          > Well, I was seeing a slowdown as well, but I was chalking it up to
                                          > the fact that I was testing RSRD LST files, and maybe they
                                          > weren't "optimized".
                                          >
                                          > I guess what would help is if you would post the .pcg and we can make
                                          > comparative measures. I can test 5.1.3 vs 5.3.2 readily.
                                          >
                                          > Andargor
                                        • Jayme Cox
                                          Here is something I posted to the developers boards recently about the rewrite off all the bonus handling code: -- As a speed comparison, I ran a load and
                                          Message 20 of 26 , Jul 30, 2003
                                            Here is something I posted to the developers boards recently about the rewrite
                                            off all the bonus handling code:
                                            --
                                            As a speed comparison, I ran a load and export of a fairly complex character
                                            from the command line:

                                            5.2 code: 34 seconds

                                            5.3.1 code: 6 minutes 2 seconds

                                            5.3.2 code: 10 minutes 8 seconds

                                            new code: 35 seconds
                                            --

                                            Speed wise the current CVS code should perform as well as the old 5.1.x and 5.2
                                            code...
                                          • frugal@purplewombat.co.uk
                                            ... Does the code cache values ? About 2 weeks ago I ran a quick profile of PCGen and loading a 12th level monk called
                                            Message 21 of 26 , Jul 31, 2003
                                              <quote who="Jayme Cox">
                                              > As a speed comparison, I ran a load and export of a fairly complex
                                              > character
                                              > from the command line:

                                              Does the code cache values ? About 2 weeks ago I ran a quick profile of
                                              PCGen and loading a 12th level monk called PlayerCharacter.getBonusTo
                                              68000 times! As the character did not change during that period the return
                                              value for each bonus should be the same.

                                              I have not had a chance to see if the code caches calculated bonuses, but
                                              it should be a relatively trivial exercise to add if it does not: Map of
                                              Bonus String => value String, map gets emptied when character gets
                                              modified.

                                              --
                                              regards,
                                              Frugal
                                              -OS Tamarin
                                            • Emily Smirle
                                              ... errr... see, the character changes as different parts of it are loaded and added to it. A naked level 12 monk is different from one with the level 12
                                              Message 22 of 26 , Jul 31, 2003
                                                frugal@... wrote:
                                                > <quote who="Jayme Cox">
                                                >
                                                >>As a speed comparison, I ran a load and export of a fairly complex
                                                >>character
                                                >>from the command line:
                                                >
                                                >
                                                > Does the code cache values ? About 2 weeks ago I ran a quick profile of
                                                > PCGen and loading a 12th level monk called PlayerCharacter.getBonusTo
                                                > 68000 times! As the character did not change during that period the return
                                                > value for each bonus should be the same.

                                                errr... see, the character changes as different parts of it are loaded
                                                and added to it. A naked level 12 monk is different from one with the
                                                level 12 attribute modifiers is different from one with a feat is
                                                different from one with two feats... etc etc etc.

                                                Unfortunate, but there you go.

                                                > I have not had a chance to see if the code caches calculated bonuses, but
                                                > it should be a relatively trivial exercise to add if it does not: Map of
                                                > Bonus String => value String, map gets emptied when character gets
                                                > modified.

                                                Doing that would consume even more processor cycles during character
                                                load, unless we can get it to ignore the bonii until loading is
                                                finished, and then sit down and do all the math.
                                                --
                                                Emily Smirle (jerril) Beware, for my Axe
                                                GMGen Princess will cut you off at the knees.
                                                Watch out for the spikes.
                                              • frugal@purplewombat.co.uk
                                                ... But once the character is loaded it should speed things up as most of the bonus calculations seem to happen moving from tab to
                                                Message 23 of 26 , Jul 31, 2003
                                                  <quote who="Emily Smirle">
                                                  > frugal@... wrote:
                                                  > errr... see, the character changes as different parts of it are loaded
                                                  > and added to it. A naked level 12 monk is different from one with the
                                                  > level 12 attribute modifiers is different from one with a feat is
                                                  > different from one with two feats... etc etc etc.
                                                  >
                                                  > Unfortunate, but there you go.

                                                  But once the character is loaded it should speed things up as most of the
                                                  bonus calculations seem to happen moving from tab to tab.

                                                  > Doing that would consume even more processor cycles during character
                                                  > load, unless we can get it to ignore the bonii until loading is
                                                  > finished, and then sit down and do all the math.

                                                  As a question: What _is_ requesting the bonus values during load? Even
                                                  things like assigning a feat to a character should not bother to look at
                                                  the bonuses to see if the character meets the pre-reqs as the character is
                                                  known to have the feat. In what situations are the bonuses needed during
                                                  character load?

                                                  I wish I had more free time to look into this at the moment it is the sort
                                                  of challenge that appeals to me ;O)

                                                  --
                                                  regards,
                                                  Frugal
                                                  -OS Tamarin
                                                • Jayme Cox
                                                  ... The rewrite of the bonus code eliminates most of those recalcs. ... There are two problems: First you have recursive BONUS statements. For instance, a FEAT
                                                  Message 24 of 26 , Jul 31, 2003
                                                    > But once the character is loaded it should speed things up as
                                                    > most of the bonus calculations seem to happen moving from tab to tab.

                                                    The rewrite of the bonus code eliminates most of those recalcs.

                                                    > As a question: What _is_ requesting the bonus values during load?

                                                    There are two problems:
                                                    First you have recursive BONUS statements. For instance, a FEAT may have a bonus
                                                    that depends on DEX. But DEX depends on race BONUSes, on PC equipment carried
                                                    (such as armor or max load), etc.
                                                    Second is that a BONUS may have a PREREQ statement in it. So, you may only be
                                                    able to use a BONUS if you are a dwarf or your DEX is higher than 13, etc.

                                                    The information contained within the .pcg file must be interpreted each time the
                                                    file is loaded. In other words, your character is built from the ground up each
                                                    time you load, just as if you were clicking around the GUI, adding levels,
                                                    feats, etc.

                                                    The combination of these two is what causes the large number of bonus
                                                    recalculations, to see if you can have the BONUS from X which has a PreReq of Y
                                                    which depends on Z (which depends on A, which depends on B) applied to your
                                                    character or not.
                                                  • Martijn Verburg (DSLWN)
                                                    Tracker organized, number will be posted by David. K TM SB ... From: Jayme Cox [mailto:jayme@reality.net] Sent: Wednesday, 30 July 2003 8:18 a.m. To:
                                                    Message 25 of 26 , Jul 31, 2003
                                                      Tracker organized, number will be posted by David.



                                                      K

                                                      TM SB



                                                      -----Original Message-----
                                                      From: Jayme Cox [mailto:jayme@...]
                                                      Sent: Wednesday, 30 July 2003 8:18 a.m.
                                                      To: pcgen@yahoogroups.com
                                                      Subject: [pcgen] [LST - BUG/TM] natural attacks: damage bonus wrong




                                                      Per the SRD (and MM p. 61),
                                                      Bite: ...deal the listed damage plus the dragon's Strength bonus...",
                                                      Claw: ...deal the listed damage plus half the dragon's Strength bonus
                                                      Wing: ...plus half the dragon's Strength bonus..."
                                                      Tail Slap: ...plus 1 and 1/2 times the dragon's Strength bonus...".


                                                      I've checked in code to CVS which fixes all the above attacks except "Tail
                                                      Slap"


                                                      Fixing the "Tail Slap" will require a .lst file change:

                                                      > BONUS:WEAPONPROF=Tail Slap|DAMAGE|STR

                                                      Since all secondary weapons only get 1/2 STR bonus to damage, the Tail Slap
                                                      currently has 1/2 STR bonus, so it needs STR bonus added.

                                                      --Jayme

                                                      PS: Please post the tracker number here, so we know it's been entered,
                                                      Thanks!




                                                      [Non-text portions of this message have been removed]
                                                    • David
                                                      782130
                                                      Message 26 of 26 , Aug 2, 2003
                                                        782130
                                                        >
                                                        > Per the SRD (and MM p. 61),
                                                        > Tail Slap: ...plus 1 and 1/2 times the dragon's Strength bonus...".
                                                        >
                                                        > Fixing the "Tail Slap" will require a .lst file change:
                                                        >
                                                        > > BONUS:WEAPONPROF=Tail Slap|DAMAGE|STR
                                                        > --Jayme
                                                      Your message has been successfully submitted and would be delivered to recipients shortly.