EPTFor example, if you ask it for mutas and 2 hatcheries, it'll give you mutas, 2 hatcheries, and a lair.
So if you want 2 bases and lair tech, remember to only ask for 1 hatch/1 lair.
SCFusion - WoL, HotS & LotV Build Order Optimizer - Page 48
| Forum Index > SC2 General |
|
Ncinerate
172 Posts
For example, if you ask it for mutas and 2 hatcheries, it'll give you mutas, 2 hatcheries, and a lair. So if you want 2 bases and lair tech, remember to only ask for 1 hatch/1 lair. | ||
|
CarbonTwelve
Australia525 Posts
On May 12 2011 21:16 graNite wrote: Isn't it possbible to "pause" for a moment to read, but start again without searching from 0? In SCBuildOrder it was possible, just not in SCFusion atm. On May 12 2011 23:02 sansalvador wrote: Do you mean stuff like "3* probe"? I actually like that. ^^ Yeah, I meant to remove that actually  That's the start of rewriting the output code. It's only doing it for Protoss with Simple output atm.On May 13 2011 03:42 Ncinerate wrote: Just a quick note for anyone fiddling with this - be careful about "hatcheries" when making builds. For example, if you ask it for mutas and 2 hatcheries, it'll give you mutas, 2 hatcheries, and a lair. So if you want 2 bases and lair tech, remember to only ask for 1 hatch/1 lair. Yep. I might change the hatchery count to "bases" so it's a bit more useful. | ||
|
michaelhasanalias
Korea (South)1231 Posts
| ||
|
CarbonTwelve
Australia525 Posts
I also played around with the indefinite goals code and I think this has improved the results for indefinite goals quite a bit. It should be a lot better at getting you more within the time limit. http://code.google.com/p/scbuildorder/downloads/detail?name=SCFusion_v0.5.exe | ||
|
graNite
Germany4434 Posts
edit: seems randomly, maybe when its not possible to reach a waypoint. | ||
zere
Germany1287 Posts
| ||
|
CarbonTwelve
Australia525 Posts
On May 13 2011 17:40 graNite wrote: v0.5 starts crashing when it runs some minutes... edit: seems randomly, maybe when its not possible to reach a waypoint. :/ I haven't noticed that. Can you provide example targets? | ||
|
ru.meta
Russian Federation88 Posts
Couple of things: - I see no reason to display "wait 1s" commands in "detailed view" as there is a time specified explicitly - I miss for some kind progress bar (it was previously in scbuildorder) to know how long current BO keeps on top - to stop searching if it looks fine. | ||
|
CarbonTwelve
Australia525 Posts
On May 13 2011 21:51 ru.meta wrote: - I see no reason to display "wait 1s" commands in "detailed view" as there is a time specified explicitly On the one hand that makes sense, however it might not be very obvious that you needed to wait if it's not displayed. I'm in the middle of rewriting the output code though so I'll play around with it and see what it's like. - I miss for some kind progress bar (it was previously in scbuildorder) to know how long current BO keeps on top - to stop searching if it looks fine. Yeah, I've been meaning to put that back in. Will try and get it done for v0.6 | ||
|
ru.meta
Russian Federation88 Posts
On May 13 2011 22:12 CarbonTwelve wrote: On the one hand that makes sense, however it might not be very obvious that you needed to wait if it's not displayed. I'm in the middle of rewriting the output code though so I'll play around with it and see what it's like. Yeah, I've been meaning to put that back in. Will try and get it done for v0.6 Great! Can you put on project page which version of sc2 patch is used? Did you thought about moving settings of game out of source so community will be able to update it as patch is coming? | ||
|
CarbonTwelve
Australia525 Posts
On May 13 2011 22:26 ru.meta wrote: Can you put on project page which version of sc2 patch is used? Yeah, that's a good idea, I'll put it in the starter info when you first open a build order. Did you thought about moving settings of game out of source so community will be able to update it as patch is coming? Yeah, it's been suggested a few times, but I'm a little concerned about the performance impact plus I just haven't had the chance to look into it enough. Besides, for the moment I think I'm doing ok at keeping up with the patches so there isn't a huge need yet | ||
|
shingbi
58 Posts
Waypoint 1 satisfied: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Waypoint 2 failed: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Will have a look at the code tomorrow. | ||
|
Cambam
United States360 Posts
On May 14 2011 04:27 shingbi wrote: I set up a build order with several waypoints (v0.5). Waypoint 1 is satisfied, but afterwards nothing happens. It just stays at: Waypoint 1 satisfied: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Waypoint 2 failed: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Will have a look at the code tomorrow. Same with me, can't seem to get past waypoint 2 even if it has satisfied the requirements. However, I let it run for a few hours and it finally reported waypoint 2 as satisfied. I only had waypoint 1 and the target, so waypoint 2 was the final waypoint. | ||
|
CarbonTwelve
Australia525 Posts
On May 14 2011 04:27 shingbi wrote: I set up a build order with several waypoints (v0.5). Waypoint 1 is satisfied, but afterwards nothing happens. It just stays at: Waypoint 1 satisfied: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Waypoint 2 failed: 4:00,00: 242M 306G 35E 23/ 26S Income: 594M 215G Buildings: 1 Nexus 2 Assimilator 2 Pylon 1 Gateway 1 Cybernetics Core Units: 21 Probe 1 Zealot Upgrades: Will have a look at the code tomorrow. The code for SCFusion hasn't been uploaded to GoogleCode yet. I need to work out how to change the branching structure for it. As for waypoint 2 failing, did you include the requirements of waypoint 1 in waypoint 2? Ie, if waypoint 1 was 1 zealot, did you have at least 1 zealot for the target of waypoint 2? | ||
|
ChickenLips
2912 Posts
Seems like a lot of work. @scroll window couldn't you put a text cursor into the text box and jump it back to the location it was before the text update? I imagine there must be other ways to remain at the same position without directly manipulating the scroll mechanism. | ||
|
CarbonTwelve
Australia525 Posts
On May 14 2011 07:20 ChickenLips wrote: Why did you decide to re-code it or have you outlined your reasons elsewhere? Seems like a lot of work. Yeah, it was a fair bit I wanted to swap to a cross platform GUI with the hope that I'll be able to release it for Mac and Linux at some point.@scroll window couldn't you put a text cursor into the text box and jump it back to the location it was before the text update? I imagine there must be other ways to remain at the same position without directly manipulating the scroll mechanism. Yeah, I tried that but getting the current cursor position was proving very frustrating. It also doesn't really guarantee it'll be in the same place as moving the scroll bar doesn't move the cursor, and also the scrollbar can be in multiple places for the same cursor position. I'll try and look at it again at some point but I did spend quite a bit of time on it already as it was frustrating me too, so I'm not sure when I'll have a better solution available. | ||
|
ChickenLips
2912 Posts
On May 14 2011 08:39 CarbonTwelve wrote: Yeah, it was a fair bit I wanted to swap to a cross platform GUI with the hope that I'll be able to release it for Mac and Linux at some point.Yeah, I tried that but getting the current cursor position was proving very frustrating. It also doesn't really guarantee it'll be in the same place as moving the scroll bar doesn't move the cursor, and also the scrollbar can be in multiple places for the same cursor position. I'll try and look at it again at some point but I did spend quite a bit of time on it already as it was frustrating me too, so I'm not sure when I'll have a better solution available. how about a listview? this would also make prettier formatting of the more complex view options and I imagine it won't fuck up the scroll position if you edit values in the underlying data structure. | ||
|
shingbi
58 Posts
On May 14 2011 07:01 CarbonTwelve wrote: The code for SCFusion hasn't been uploaded to GoogleCode yet. I need to work out how to change the branching structure for it. As for waypoint 2 failing, did you include the requirements of waypoint 1 in waypoint 2? Ie, if waypoint 1 was 1 zealot, did you have at least 1 zealot for the target of waypoint 2? Ah no, I didn't. I guess it's because of the new "1 to 2" syntax? So if I have zealot in my second waypoint set to 0, but I already have a zealot, it's going to run into trouble? I can see why it is that way... but it's also kind of a pain.  I mean, if the optimizer puts in an extra zealot and still hits all the targets... I'm not gonna complain, now am I? Maybe do two columns (from/to) and set the "to" column to max by default?Would be cool if you could put the source for the new one up. Are you using Qt or GTK or something completely different? | ||
|
CarbonTwelve
Australia525 Posts
On May 14 2011 19:17 shingbi wrote: I guess it's because of the new "1 to 2" syntax? So if I have zealot in my second waypoint set to 0, but I already have a zealot, it's going to run into trouble? Yup. I can see why it is that way... but it's also kind of a pain. I mean, if the optimizer puts in an extra zealot and still hits all the targets... I'm not gonna complain, now am I? Maybe do two columns (from/to) and set the "to" column to max by default?I agree it's a pain. The plan is to change it so that waypoints will always include the earlier waypoints, so that if you put 1-2 zealots for waypoint 1 then waypoint 2 has to have at least 1-2. Would be cool if you could put the source for the new one up. Are you using Qt or GTK or something completely different? It uses wxWidgets. I looked at Qt but had issues trying to use its property grid (part of the trouble is the property grid isn't part of the base Qt widgets). I've got a feeling I looked into GTK+ but rejected it for some reason, can't remember why... Either way, wxWidgets seems to be fine atm. Plus I like that it uses the native controls so that it looks like a native app for whatever platform it's built for. | ||
|
Cambam
United States360 Posts
It'd be nice to have an option so that makes it so the program won't have an infestor finish unless pathogen glands is finished. Most people do this by not starting their infestor production until the pathogen glands upgrade is at least 31 out of 80 seconds completed. Infestors take 50 seconds to morph, so they pop out right after the research finishes. I'd like to see higher defaults on the hatchery and extractor (maybe 1 to 6 and 0 to 10 respectively) Keep up the great work. | ||
| ||
