Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
PEMDAS
Parenthesis then exponents, then multiplication, then division, addition, then subtraction.
The right calculator is the correct one.
EDIT: Further, the parenthesis is not separated from the 2 by a multiplication symbol thus the parenthetical portion is assumed to be directly associated with the 2 before the 2 is associated with the division symbol.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
Jerrietheferry wrote:
It would be really awesome if somebody could make a any% TAS of Commander Keen 6 on easy difficulty. I believe the fastest TAS of the game at this point is that of Turska in 3:20.25 on hard difficulty. Since the any% category of this game uses easy difficulty, it would be nice to see how fast a TAS could finish it.
Also, I speedrun Commander Keen 6 any% myself and my PB is 3:23.19 (with some major mistakes), and my sum of best is a 3:19 (with manual splitting so not 100% accurate). So a sub 3:20 should definitely be possible for a TAS!
I don't have any experience with making a TAS, but perhaps someone here is up for the challenge!
Generally the TAS standard is to use the highest difficulty setting available. Unless there's reason for using a lesser difficulty (different game-content not seen at higher difficulties), it's frowned upon to use an easier setting simply to yield a marginally faster time.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
petaQ wrote:
...not to mention all that loading time on a TAS would be fugly.
While tapes and disks do lead to long loading times and inflated TAS times, it's understood and isn't likely to impact the judgement of any submission. See this run for an accepted example with a long loading time.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
petaQ wrote:
C64 4 in 1 game cart doesn't work currently, unpaused Bizhawk auto-starts the first game, failing that you can use Joystick port 2 on frame 4 to select the second.
Member J posted
"Can confirm this behaviour with up to date developer version of BizHawk. Works fine on Vice and CCS64.
Fiendish Freddy's Big Top o' Fun + International Soccer + Flimbo's Quest + Klax (Europe) (C 64 Games System).crt
SHA1: 0b35ea869239ed1edbc4b31e6d5da185bf88ea19
With TAStudio, there's one frame (#4) where you can press down on Joy2 to select the second game. But things don't seem to work further than that. Moving around in TAStudio (after adding the Down on frame 4) also breaks things further, need to reboot core every time for it to run expectedly. "
I'm hoping to Tas Flimbo's Quest having recently worked out how the rng functions. I'll try tape images meantime, I can't find a different compatible cart file.
I have no suggestion for this problem, but.....IIRC, using multi-game carts is frowned upon unless it's not possible to find the game you're wanting to TAS by itself.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
feos wrote:
Absolutizing things does no good...Overall, I don't think there's reason to blow things out of proportion here.
It was not my intent to blow things out of proportion. I did not intend for any of these posts to be interpreted that I am upset with the current system. I'm not...unless the example I used (judges completely disregarding the community in preference of their own opinion) is actually happening. Forgive me if my choice of words did not convey this or came across too harshly.
ThunderAxe's post that I first referenced can be interpreted as though votes don't matter to judges unless they are explained with an accompanying forum post. I don't believe this is currently happening. I only want to re-enforce the idea that vote totals should matter to judges when determining tier. I recognize that they aren't the only factor considered.
I wasn't suggesting, nor am I now, that any of the site's current (or past) judges are guilty of this. I've been quite pleased with how things have been judged since I've been around. If there is a current judging problem, I'm unaware of it.
I'm not trying to start (or prolong) a debate here. Yet, I stand by my opinion that judges should not go against the general community opinion simply because they personally disagree with how entertaining a submission is. I'm not saying this is occurring, just that it's something I believe shouldn't be.
Side Note:feos wrote:
Second, millions of flies can be wrong.
I found this statement hilarious, assuming you meant that millions of flies will still swarm around s**t even though it's s**t. Great reference!
Point considered (along with some statements in IRC). Is the voting then worthwhile in the first place?
Maybe a new policy on determining community entertainment value would be to only use forum posts. We could eliminate the voting mechanism and simply put our yes/no vote in the forum for a given run. While this would unfortunately inflate the length of many topics, it would ideally lessen the abuse.
EDIT;
I guess the point I was trying to make in my earlier post is that a judge's personal opinion should never be able to override the collective opinion of the community (however that collective opinion is determined).
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
ThunderAxe31 wrote:
Also, submission votes are not the only factor for a judge decision.
If you're saying that submission votes aren't the only factor in whether a submission is accepted for publication or not, I agree with that.
However, if you're claiming that votes aren't the only factor on what tier a submission is accepted to, I respectfully disagree with your statement. In my opinion, a judge's personal opinion on whether a TAS is vault or moon quality should not override the community voting.
If the voting is borderline, I have no problem with the judge being the determining factor for the resulting tier. If the voting is clear on how entertaining others find a submitted run, yet the judge's opinion is opposite; the judge should not be allowed to override that voting and put the run in the tier of their own personal perspective.
This applies in both directions. A judge shouldn't be able to override a low entertainment vote to promote a run to moons, nor should they be able to override higher entertainment vote to relegate a run to vault.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
slamo wrote:
I just had a frustrating couple of hours trying to get some of these games to work and I feel like I should document this somewhere. There are quite a few games that will not run in JPC-rr and won't be TASable anytime in the near future. All of these games either throw processor exceptions or an error message about "int 15h" not being supported...
I've been playing around with getting CD games functioning in JPC-rr.
I've created a new FreeDOS boot image (HDD instead of floppy) that installs the CD-ROM driver.
Then I made a CD image of Warcraft (well the data portion at least--JPC doesn't support CD-DA discs).
I then assembled JPC with the following settings:
*The new FreeDOS boot image mounted in the hda slot
*A blank 256mb disk image mounted in hdb slot
*the warcraft data CD image mounted in the CD-ROM slot
*changed boot device to hda instead of fda
I was able to perform a full-install of the game from the CD to the blank drive and run it.
Further, I was able to then dump the drive with the installed game on it to a new disk image and then re-import that dump for a new mountable game image.
I can run the game from this new game image as well (even using the floppy FreeDOS boot image) as seen below:
Link to video
Sorry about the odd sound (it may be resulting from using a different boot disk than what I installed with).
For what it's worth, the console reports the "error message about "int 15h" not being supported" both with the installation program and starting the game itself; but all seemed to work fine thus far.
I don't yet know about games that need to access the CD drive for data during gameplay, but this is at least a good start.
I just wanted to show that the "15h error" may not be a death sentence for some games.
In my opinion, this may be a big step toward getting many other games to work with JPC-rr.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
MAJOR UPDATE (at least in my opinion it's major):
I've been playing around with getting CD games functioning in JPC-rr.
I've created a new FreeDOS boot image (HDD instead of floppy) that installs the CD-ROM driver.
Then I made a CD image of Warcraft (well the data portion at least--JPC doesn't support CD-DA discs).
I then assembled JPC with the following settings:
*The new FreeDOS boot image mounted in the hda slot
*A blank 256mb disk image mounted in hdb slot
*the warcraft data CD image mounted in the CD-ROM slot
*changed boot device to hda instead of fda
I was able to perform a full-install of the game from the CD to the blank drive and run it.
Further, I was able to then dump the drive with the installed game on it to a new disk image and then re-import that dump for a new mountable game image.
I can run the game from this new game image as well (even using the floppy FreeDOS boot image) as seen below: The mouse movement is weird on this because it uses the old floppy FreeDOS without the corrected ctmouse command.Link to video
Sorry about the odd sound (it may be resulting from using a different boot disk than what I installed with).
I don't yet know about games that need to access the CD drive for data during gameplay, but this is at least a good start.
In my opinion, this may be a big step toward getting many other games to work with JPC-rr.
Still can't get Star Trek to work though....but I'm not giving up yet.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
UPDATE:
I've successfully created a FreeDOS boot image (based on the most recent FreeDOS 1.2 release) that automatically recognizes the CD drive for JPC-rr. As an added bonus, it loads ctmouse with the /r32 linear settings by default.
There are a few catches:
*The boot image is a HDD image, not a floppy image.
*Unfortunately, I still cannot get Star Trek to work properly.
Any progress is progress!
I'm going to try some other CD based games and see what I can accomplish.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
I was able to contact Ilari on IRC.
ST 25th Anniversary uses CD-DA for the music. CD-DA is not emulated in JPC-rr.
So even if we could use the CD version of the game, it wouldn't help with the music.
He also said that the problem with getting the CD to work in the first place is that most of the floppy images of FreeDOS used for booting don't contain CD-ROM drivers.
I'm looking into seeing if I can make a boot image that would offer this functionality.
I'll keep you guys updated.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
You guys are awesome. I love the collaboration here.
Arc wrote:
The extra point attempts can be completed faster by running forward and letting the other team tackle the kicker.
I actually figured this one out myself while testing out the stuff mentioned above. I've already updated my run,I just need someone with power to update the submission file. EDIT: Thank you, Mothrayas END EDIT
Running out of bounds did yield faster halves. As did getting tackled instead of extra points. This extra point strat specifically allowed for a major reduction of frames (that I could have done before if I had thought about it); I stopped input mid way through the kick return on the 2nd half of the super bowl, cutting out a TON of frames.
The run will still play through to the 'game over' screen. The score total is just smaller due to missed PAT opportunities.
I'll be updating the submission text and temp encode asap.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
hopper wrote:
So, here's what I've decided. I'm going to post my findings here. If you're willing to learn the craft and make a high quality TAS with that information, you can make your run and have the glory of having the first published Yars' Revenge TAS. The clock will be ticking, though. If you're unwilling or unable to make a TAS to this site's standards, I'm going to start working on mine very soon.
Donnaken15, if you indeed use hopper's info to complete a TAS, be sure to give credit for that info in the submission notes.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
Kenttsu wrote:
Early issues:
1) I haven't been able to get the sound blaster to work. I'm getting this message:
Does the game work with any other sound settings, or do they all produce this error? (obviously works without sound).
Kenttsu wrote:
2) It's one of those games that run notably faster with a faster CPU. Despite being a late era DOS game, I feel that the default 20MHz CPU speed is optimal for this run as it makes the run still relatively easy to follow despite being quite fast (with the fastest in-game speed setting).
New processors in 1995 averaged in the 150-200mhz range (see here). Even for the few years before that they were well above 20mhz. I'd bump the CPU speed up myself. This can be done by decreasing the CPU divider during assembly in JPC-rr. A CPU divider of 5 will yield a 200mhz system; 6.6.... will yield a 150 mhz system (I don't remember if you can do decimals in the CPU divider setting or not).
While the resulting video at this speed may be incomprehensible a slowed down version can be encoded as well to watch what's actually happening while still getting the shorter time for the completed TAS itself.
Kenttsu wrote:
Early observations and strategies:
-Levels can be viewed in game's editor. I'll post pics later.
-Before starting the game I set the game speed to 100% and number of players to 1 (otherwise it would start a death match mode).
-Item boxes contain random amount(?) of random items. Box's content is decided when player picks it up so RNG manipulation will be really important as some items are very expensive and can be sold.
-Getting enough money for good digging equipment and armor is probably worth it
-Player can buy teleport items (round hole) that can be used to move to the teleports that are already on the map (the second level in the test run skips half of the level)
-The fourth level and some of the later levels got several exits. But it appears that only one is available. I haven't yet tested if the exit location is random.
Use RNG manipulation as efficiently as possible to yield your best box contents within reason for frames lost to manipulation.
Look into the teleport stuff more. You may be able to cut a lot of game out using these.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
phoenix1291 wrote:
Probably a silly question, again, but how to use fastloaders?
I asked about fastloaders back in July. SaxxonPike tested a few with the core and said that some worked, but I don't believe specifics on how to use them were detailed.
I had suggested it might be implementable as an optional part of the firmware setup (assuming this is possible from a coding perspective).
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
Chamale wrote:
Why is it faster to use RTA strats after the first three halves? Is it just that it's not possible to outmaneuver the AI defenders without releasing the Up button? Or is it related to the clock having only 30 seconds on it?
Correct on the timer. I tested through every half until it stopped being beneficial to run a regular play; after the first 3 halves.
Also, did you test running out of bounds instead of running directly into the defenders? For example, in High School first half, that stops the first play at your 41-yard line instead of your 44-yard line, so it's slightly sooner and will let the clock run down a bit more.
I did not consider running out of bounds.
I may look into this, but it will have to wait. I'm focused on other projects at the moment.
I'm guessing that going OOB instead of being tackled will change the RNG of what formation the defense chooses. If so, all the dodging will then need redone from scratch. If not, it may be easy to update and re-sync by simply adding L or R to go OOB then add a few frames for the extra few yards to reach the endzone.
EDIT: Ok I had a few brief moments to look at this. Running out of bounds does indeed result in a different defensive formation. I didn't yet have time to TAS in all the dodging to see if it indeed yields a faster half or not than in the current submission.
As the defense is definitely a different formation it will require at least 2-3 halves of completely redone TAS work if it is indeed faster. The reason I say 2-3 is that in the first half of college, the extra distance to run might fully run out the clock and result in a slower half than running the play as currently submitted.
I will test further as I am able. Leaving the current submission for now.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
MUGG wrote:
About the read+write thing, my problem was as follows:
I lost my leading savestates, so I played back the movie and got new savestates.
But when I loaded those states, the emulator would play until the movie's original end each time.
I deleted the movie file and the savestate's backup file, but it didn't fix the problem.
The "Truncate" option worked for me though.
"Truncate event stream" simply removes all input events beyond the current point in the loaded movie.
Savestates will always have all their input events. When loaded as a movie instead of a savestate, it will play to the end of the savestate's event stream. The savestate itself will retain its event stream even if you truncate the currently running movie. To change the savestate, overwrite it with a new save after truncating.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
I've made some progress.
Within jpc-rr, I edited the fdconfig.sys file to contain the line
device=emm386.exe RAM
I then had to edit the autoexec.bat file and delete
cutemouse
I reset the system (from the file menu).
Then tried to run the game.
This got me past the memory problem....but now freedos doesn't have the CD drive installed....I don't know how to do that part.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
MUGG wrote:
When creating an image, JPC only lets me choose between floppy and hard disk. It does not let me choose CD. (If that makes sense.)
When importing a disk image, select your CD .cue/.iso file as your 'Image file/Directory'
It should then add 'CD-ROM Disc' to the 'Image Type' drop-down box.
EDIT:MUGG wrote:
@DrD2k9
I tried to run an HD image of the (trimmed) CD version.
It tells me the game needs 560K of free low memory and 1024k of free EMS (expanded) memory.
I currently have 610K free low memory and 0K free EMS memory.
How can I set up enough memory to run the image?
Try not skipping autoexec.bat at startup. That should load the needed memory.
There's also slight possibility that memory amount in the PC settings is actually set too high. This can occasionally cause some games to think there's not enough.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
MUGG wrote:
How can I make sure the game gets enough cycles in JPC-rr?
You can set the CPU speed by adjusting the 'CPU Divider' during assembly. A higher number yields a slower emulated CPU. Divide 1 GHz by the divider number to get the effective CPU speed.
As this game was released in 1992, the default CPU speed in JPC-rr should be fairly accurate already.
As a note: the faster the emulated CPU speed is set, the slower it will appear to run while TASing or playing a movie in the emulator itself. The dumped AVI will play in a media player at the appropriate rate for the CPU speed settings.
I hope all that made sense.
Also, if I want to run the CD version, and I have the game as a .bin and .cue file, how should I proceed to run it in JPC?
I've not yet done a CD based game that required the CD to play. But I believe you can make a JPC-rr image of the CD as you do an HDD image of the installed game.
In the assembly settings I think you can set the CD drive image similar to setting the HDD image.
c-square or slamo may have better info on this.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
MUGG wrote:
I can select neither Save>Movie nor Start Dumping while I'm running the game. I can save movie before "starting" but when I "load" it later, it doesn't seem to play back.
Stop/pause the game (T) and both of those options will become available.
When you save a movie, it essentially makes a file similar to a savestate with all inputs up to the point where you save the movie. The difference between the files is the savestate file has extra information; the movie file will have minimal necessary info and input events.
Savestate files and movie files are the same file type just with slightly different information. You can open/playback a savestate as a movie file.
Once you have a saved movie file/savestate with inputs; load the movie and start dumping before starting the movie. Once the movie runs as far as you want to dump, you can stop the movie then stop dumping (also, I believe it's possible to simply close the emulator and the dump will be good).
Once you have a dump file then you can convert that to a temporary encode (avi, etc.) that you can watch on VLC or other appropriate video player.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2088)
Joined: 8/21/2016
Posts: 1026
Location: US
c-square wrote:
DrD2k9 wrote:
One note on this. Some programs accelerate the mouse movement in weird ways, so clicking and dragging doesn't always drop the cursor exactly where you stop.
This could be for two reasons.
First, ctmouse by default accelerates mouse movement, meaning that moving x by 100 has much more than double the effect of moving it by 50. You can turn off this setting by unloading ctmouse and reloading it using "ctmouse /r32". That will make it scale linearly, meaning 100 will be exactly double the movement of 50.
To check if this is the case, use the 'M' key to bring up the manual mouse movement screen and move your mouse's x by 10. Reload and move by 20. Reload and move by 30. If it's linear, then it should move equal distances between the three. If it's not, then 30 will be much bigger than three times 10.
Second, if the resolution of the mouse area is different than the games I've been playing, then it will move differently. You can adjust the sensitivity of the mouse movement by using the 'Y' key to bring up the mouse speed screen and alter the numbers till you find ones that work. Note that this will only work if the mouse is moving linearly.
Thanks! I'm pretty sure mine wasn't running linearly. Glad I wasn't very far into the Indy run. I'll try the different settings.
