|
I'm pretty sure that the layered control groups are wrong on the spreadsheet but they seem to be correct on the visualizer as well as the hotkey setup within the game.
My Two Cents:
As for me, I've been using TheCore for about 2 weeks and there are both good and bad points to the system. Sometimes I think TheCore is trying to be as unintuitive as possible. However, there are things you can do with TheCore that you cannot do otherwise with a traditional hotkey system. I play at a relatively high lvl (mid masters) so finding something that actually makes me play better is important to me.
At first I started with ZRM (Zerg, right hand) on qwerty. It was pretty uncomfortable at first, but I decided to run with it so I see how well the system would work out. After the two weeks were up, my hands were aching. It's natural to want to curl your fingers into your hand, but quite a few of the keys to create units or building were downright difficult to reach. Such as make Ultralisk or make Spawning Pool. Using the abilities for the units seem almost haphazardly binded. Even the research upgrades don't make much sense to me. I managed to memorize most of the abilities for the units, but the upgrades I had to change and I switched them to a grid layout.
I made many other major changes to TheCore as well. So I guess you can say my Core is a extremely modified one. I made the switch to ZRS. That was a blessing to my hand. Many of the keys I could not reach before, I could now reach. But I'll continue referring specific hotkeys to the ZRM JIOP setup for standardization sake.
I don't know why burrow and unborrow are on separate hotkeys. The only reason in my mind I believe would be that if you thought you selected a group and some units are burrowed and the other half is unburrowed and you wanted to burrow or unburrow them, you could do that. Turns out, you cannot do it anyways. The burrow and unborrow option are in the same slot on the command card. So it makes little sense to try to put them on different hotkeys, so I made them the same. This also applies to burrow research and "burrowing" spine crawlers and spore crawlers. Moving a spine or spore with an ability button seems too unintuitive.
The burrow hotkey is placed a little to close to the attack button. Sometimes I would be in a large scale team battle and I would miss the attack button and push the burrow button instead. Then I would look like the fool because I needlessly burrowed my entire army mid battle and lose the game. I moved the burrow button to a different but still easy to reach location. The patrol hotkey was also in a difficult to reach space, so I moved it to the left one slot because I use it rather heavily.
There are many positive sides to TheCore as well. Shift injecting and camera recall to specific bases for example. However, you MUST have all your control groups set up properly in the first place or it won't work. Things are a breeze when you have things set up, but sometimes you forget and binding new units to the control groups takes time. For example, when you're on one base, you have to bind your main hatch, bind your queen, then bind your camera. That is 3 binds. When you have a natural, you have to bind your new hatch, bind your new queen, and bind your camera again. That is 6 binds total. Continue adding bases and that is a lot of extra binding. With my previous setup, I only had to bind the hatches and bind the queen to the hatch. Meaning only two binds per base.
I'll continue experimenting with alternative hotkey locations, but the current Core doesn't quite cut it for me. After making my changes though, I've played noticeably better.
|
On May 28 2013 10:37 Tritone wrote:
So, injecting queens should be on Shift + W, and hatcheries should be on (no shift) W, right?
Thanks a lot for your help. I'm 95% sure I'm gonna switch to this. The standard for lefty keys don't work well for me for zerg.
That is correct.
One thing I noticed: instead of ctrl/ctrl+shift+R for add to/create inject control group, it should be on ctrl/ctrl+shift+F
|
On May 28 2013 11:15 daagar wrote:
What is the suggested layout? I realize the obvious answer would be 'random', but my fear is learning it and realizing down the road that I want to main Zerg afterall and having to relearn... would it be better to use the Zerg layout even when playing P or T, swap between layouts based on race, use P or T and 'suffer' with Z, etc?
I believe (someone told me this but I never checked) that the Protoss, Terran, and Random layouts are all exactly the same.
So there aren't 3/4 layouts- there's really only two.
I decided that I'd just get used to switching layouts, but I made a few minor adjustments. For example, in TRM stop is 8 and hold position is K, whereas in ZRM stop is K and hold is 8. I wanted to make sure those keys were the same in both layouts, so I changed TRM to be like ZRM.
However, I haven't really played Terran at all yet except for a game or two against the computer to test out the hotkeys, so I can't say yet whether or not learning two layouts is more trouble than it's worth.
One thing that might be worth mentioning is that Fengshaun, (the guy who said that playing zerg with the random layout wouldn't be a problem) went through the same dilemma about which layout to use. I believe he eventually settled on using the Zerg layout, because of how it has an extra homerow key for abilities (O).
If I were using the ZRM layout to play protoss/terran, I'd consider ignoring the suggested control groups a little bit. Rather than having CCs/Nexus on P and unit production (rax, starport, etc) on shift+P, I might use 0 and - (or maybe P and 0), so that it would be easy and fast to "tap" those control groups to check on your production.
|
I have a problem with larva injects with the new TheCore layout, 0.7. I used to be able to hotkey my queens to shift+P by pressing ctrl+J, but now I cant in the new version aside from the 0.6.3. In the google datasheet, it tells me to bind queens to P with the hatcheries, but then I cant macro as easily as I have to press P, then click on the hatchery icons, then spend my larva. So I'm clueless to how I should hotkey my queens. The layered cameras work fine though.
|
Right now it's control k. (Instead of j. Control j is a camera.)
Soon, it will likely be switched, as I believe we are in the process of making things a touch more efficient. Jak's on vacation, so everything is pretty much in stasis till he gets back 
|
Cool...Also on the spreadsheet its Core 0.7, but the one I'm using now is the Beta version. Are those two the same thing??
|
On May 29 2013 01:29 Lseraphim wrote:
Cool...Also on the spreadsheet its Core 0.7, but the one I'm using now is the Beta version. Are those two the same thing??
Yep, since these keys are expected to be so close to the final layout.
|
On May 28 2013 15:22 Bulgogi wrote:
I don't know why burrow and unborrow are on separate hotkeys. The only reason in my mind I believe would be that if you thought you selected a group and some units are burrowed and the other half is unburrowed and you wanted to burrow or unburrow them, you could do that. Turns out, you cannot do it anyways. The burrow and unborrow option are in the same slot on the command card. So it makes little sense to try to put them on different hotkeys, so I made them the same. This also applies to burrow research and "burrowing" spine crawlers and spore crawlers. Moving a spine or spore with an ability button seems too unintuitive.
Alternating between two keys is much faster than pressing one key repeatedly, which is sometimes what is necessary to eliminate the discrepancy between burrowed and unburrowed units, as you stated.
On May 28 2013 15:22 Bulgogi wrote:
The burrow hotkey is placed a little to close to the attack button. Sometimes I would be in a large scale team battle and I would miss the attack button and push the burrow button instead. Then I would look like the fool because I needlessly burrowed my entire army mid battle and lose the game. I moved the burrow button to a different but still easy to reach location. The patrol hotkey was also in a difficult to reach space, so I moved it to the left one slot because I use it rather heavily.
I don't see how this is a problem. My pinky is almost always fixed on J (K) because there are so many abilities on that key. I only move it for the little-used H, N, and M keys. (J, M, and Comma) Actually, I would imagine that Zerg uses these keys less often than Terran because three of the home keys are abilities compared to Terran's two.
To separate this argument from my personal tendencies, you would probably get used to it. Two weeks is not enough to adjust completely.
(Emphasis on "probably")
On May 28 2013 15:22 Bulgogi wrote:
There are many positive sides to TheCore as well. Shift injecting and camera recall to specific bases for example. However, you MUST have all your control groups set up properly in the first place or it won't work. Things are a breeze when you have things set up, but sometimes you forget and binding new units to the control groups takes time. For example, when you're on one base, you have to bind your main hatch, bind your queen, then bind your camera. That is 3 binds. When you have a natural, you have to bind your new hatch, bind your new queen, and bind your camera again. That is 6 binds total. Continue adding bases and that is a lot of extra binding. With my previous setup, I only had to bind the hatches and bind the queen to the hatch. Meaning only two binds per base.
Agreed here, although I particularly dislike inject as a whole as it is one of the main reasons I don't play Zerg.
Actually, I also experience this when hotkeying my army. Although pros usually do it in a similar fashion to myself, as I've seen on stream, I just find it annoying to assign units to different control groups so often. Hopefully I'll get used to it. 
On May 28 2013 15:22 Bulgogi wrote:
I'll continue experimenting with alternative hotkey locations, but the current Core doesn't quite cut it for me. After making my changes though, I've played noticeably better.
Great for you.
As a closing note, I am not criticizing your changes, but rather criticizing your arguments. Personal experience always trump statistics because nobody is perfectly normal. However, I also believe it is wrong to make the judgement that everyone shares the same problem.
|
Hey Anty,
I think I'd like to disagree with you, and I'd love for you to tell me why you think I'm wrong. I don't think personal experience always trumps stats, and I'd argue that it rarely does, at that. Statistical, aggregate data is far more important than "clinical" data in medicine, psychology, business, economics and likely in most fields, for the simple rule that one is actually more likely to be closer to the mean than to the outliers. You are right that no one is perfectly normal, if you define normal as a particular point on a curve. If you let "normal" be a range, by definition, most people are normal.
I bring this up here, because I think a philosophy many people have is "I'm going to try theCore, and then anything I immediately don't like, I'm going to change. After I change it, I'll like it better, so it probably is better." Now, above you said that you believe that mentality is wrong, and I think you were spot on. But I also think that mentality is wrong to apply to oneself, without significant testing. (By significant I mean months of practice not only with TheCore, but with changing different variables one at a time). The reason for this is that unless one can identify a systematic error in Foxy's and Jak's methodology (TheCore is not perfect science after all, so my point is very much weakened by this fact), it is unlikely that one is radically far from the mean even if it feels like one is. It will naturally feel that way, because the layout was designed for efficiency, not learning. In general, almost every function did not take into account learnability (notable exceptions are some upgrades), so one can induce that it will be tough to learn and not feel right. In particular, when someone says "I make the mistake of hitting X instead of Y sometimes, so f(X) needs to change", that raises a huge red flag: it's more likely that one's fingers have not learned how to distinguish between X and Y, than X really being a bad location for f(X). (Sorry for the mathiness, I hope it's clear that by f(X) I mean whatever cg or ability is on X.)
Anyways, I'd love to hear your thoughts on this, because I always get confused when people modify things because it feels better, at least without a lot of experience with TheCore. (If one modifies based on their personal key scores, then my comment does not apply to them).
@Bulgogi,
TheCore is not directly trying to be unintuitive, but it does have that effect. One of the very lowest desiderata (things one may want in a hotkey layout) according to TheCore philosophy is learnability or intuitiveness. Upgrades are actually standardized, though it may be hard to see that. For Zerg, missile and air attack are the same, both defence upgrades are the same, and melle and upgrade spire are the same (cuz there is no air melle but broods do melle damage, sort of).
Re: burrow. The reason they are on different keys is basically because of what Anty said above. Also, one can do really cool micro by holding down the burrow key and clicking units. This is impossible if burrow and unburrow are on the same keys. I'm not sure what you mean by the fact that it is unituitive moving spines and spores with a button. As for burrow being near attack, I think that is a matter of practice. Anecdotally, I used to do the same thing, and now I don't.
You are right that injecting and the whole setup takes a lot of getting used to.we are in the process of making it a touch easier. Basically:
Cg 1 is hatches, cg 2 is queens.
Add to cg 1: alt j
Select cg 1: p
Create cg 1: ctrl shift j
Add to cg 2: ctrl p
Select cg 2: shift p
Create cg 2: ctrl shift p
This has the effect of making it easier to add hatches into cgs and make base cams faster. Further, queen control group gets "normal" in the sense that it is added to with the same button that it is recalled with. (One may argue that because one needs to Center a macro hatch, this is not worth it. I have a good argument against this, but it is long to type out. Ill happily do it if requested though.)
Thanks a lot for the feedback. I agree with Anty that your arguments aren't very strong, but the comment is very much appreciated. As Antylamon said, definitely do your own thing. Just, IMO, don't do it too fast
|
On May 29 2013 06:02 Ninjury_J wrote:
Hey Anty,
I think I'd like to disagree with you, and I'd love for you to tell me why you think I'm wrong. I don't think personal experience always trumps stats, and I'd argue that it rarely does, at that. Statistical, aggregate data is far more important than "clinical" data in medicine, psychology, business, economics and likely in most fields, for the simple rule that one is actually more likely to be closer to the mean than to the outliers. You are right that no one is perfectly normal, if you define normal as a particular point on a curve. If you let "normal" be a range, by definition, most people are normal.
The problem is that TheCore is not a range. Yes, there are hand size differences, but that's nothing. Personalization creates that range, and that's the irony of it all.
TheCore is entirely based on the statistically most-used commands, as you know. There are, of course, pros to this. The most notable instances are that of the calculation of available keys to find the home keys and the finger stress calculations to quantify priority. The primary con I can think of is the first instance I can think of (besides hand size) in which variables show up: the number of times each command was used. This only applies to half of the layout, but half is still a lot. That half includes every ability and every unit in the game. The usage of these commands varies tremendously, from PDD vs HSM to Repair vs Return Minerals. Then there's the big ones, which are unit production structures (especially larvae) and the build structure command. The variation here is obvious, so I don't think I have to elaborate here.
I still haven't even touched on base cameras. For Terran and Protoss especially, you're rarely ever going to have to use 5th or 6th camera keys. By that time your main and nat are almost guaranteed to be mined out if your name isn't Dragon, so you can just use your production camera. So if the 5th and 6th base cameras are relocated to, for instance, N and K (a la J and I's cameras), then you're left with the Shift+L and Shift+Semicolon, which could potentially be layered control groups, next and last subgroup, etc.
I could go on and on about how customizable TheCore really is. I think you are underestimating how much little changes can stack up to make TheCore much less useful, especially if Ctrl and Alt are relocated. But Ctrl and Alt are a story for another day. (A story hopefully told to Blizzard staff)
On May 29 2013 06:02 Ninjury_J wrote:
I bring this up here, because I think a philosophy many people have is "I'm going to try theCore, and then anything I immediately don't like, I'm going to change. After I change it, I'll like it better, so it probably is better." Now, above you said that you believe that mentality is wrong, and I think you were spot on. But I also think that mentality is wrong to apply to oneself, without significant testing. (By significant I mean months of practice not only with TheCore, but with changing different variables one at a time). The reason for this is that unless one can identify a systematic error in Foxy's and Jak's methodology (TheCore is not perfect science after all, so my point is very much weakened by this fact), it is unlikely that one is radically far from the mean even if it feels like one is. It will naturally feel that way, because the layout was designed for efficiency, not learning. In general, almost every function did not take into account learnability (notable exceptions are some upgrades), so one can induce that it will be tough to learn and not feel right. In particular, when someone says "I make the mistake of hitting X instead of Y sometimes, so f(X) needs to change", that raises a huge red flag: it's more likely that one's fingers have not learned how to distinguish between X and Y, than X really being a bad location for f(X). (Sorry for the mathiness, I hope it's clear that by f(X) I mean whatever cg or ability is on X.)
Yes, I am familiar with f(X) as function of X.
I agree with a lot of this, but it seems that you are directing this at Bulgogi's side of the argument rather than mine, with the exception of: "The reason for this is that unless one can identify a systematic error in Foxy's and Jak's methodology (TheCore is not perfect science after all, so my point is very much weakened by this fact), it is unlikely that one is radically far from the mean even if it feels like one is."
While I can vouch for the underlying concept of this sentence, I abide by the contrary opinion that the bell curve gets steeper more quickly than you think. At the very top of the bell curve is the person who is the absolute perfect match for TheCore. Nobody, of course, matches TheCore perfectly. The only problem is that not everyone realizes it. At one point of the bell curve, which I will call the event horizon, everyone farther away from the norm than that point should not choose to use TheCore. Among these people are extremities such as people who play on laptops and people with disabilities. Almost directly on the event horizon, much more average people are present. Most importantly, they have the ability to use TheCore with the same physical capabilities as the person at the top, but they have an alternative which is notably more efficient than any close resemblance to TheCore. Base FGHJ is one of these, even though it is a fair distance closer to the norm because rebinding Ctrl and Alt is usually not tournament legal without rewiring your keyboard. (If I am not mistaken, most tournaments require you to bring your own keyboard. I know for a fact that MLG does.) Base NJIO is another possibility. Remember the key fact that most people do not think of these options, not to mention realize that it could be better for them personally. Just as important is the fact that base NJIO is almost certainly more efficient than JIOP for some people and vice versa.
Oh, and if I am not mistaken, camera keys are not trackable, so it is impossible to reliably quantify how often the 5th and 6th base cameras are used. Volunteer participants would be required and unfortunately we do not have a deep list of pros who would be willing to volunteer for Jak. People frequently associate him with the Dunning-Kruger effect. Anyways, that's one more uncertainty which backs my opinion.
|
I actually agree with you largely. TheCore is a point, thanks for pointing that out. Still, I think that without evidence, changes based on feeling are likely to hurt more then help. I.e. if one makes a change without numerical backing (or at least really really good logic) one is more likely to move farther away from one's potentially perfect model than towards it. Granted, I don't think that difference in probability is very high, and it's probably true that for most players, feel really is a decent heuristic for what's best for them. Perhaps my calibration on TheCore's suitability versus people's ability to reliably self-modify is too high. (I'm not putting down people's modification ability in general, I just find it unlikely that personal opinion/anecdote is more likely to be correct than TheCore-as-is, without evidence). That being said if one is a high level player and has a lot of in game experience, they may know ways to tweak TheCore for their style. Or, sometimes people are down right genius, after all, that's how most changes get implemented. For the record, I'm not saying that people should take TheCore as is like Dogma (I've critiqued it myself since the beginning), I think debate, critiscm and experimentation are part of what makes TheCore so practical. Im just of the belief that most changes that are good for one (baring things like changes based on ones hand) are likely good for most people too. Thanks for the reply
Question: if you could layer control groups, (I.e. switch l and semi for layered cgs), what would you put on them? What modifications have you made?
|
On May 29 2013 08:33 Ninjury_J wrote:I actually agree with you largely. TheCore is a point, thanks for pointing that out. Still, I think that without evidence, changes based on feeling are likely to hurt more then help. I.e. if one makes a change without numerical backing (or at least really really good logic) one is more likely to move farther away from one's potentially perfect model than towards it. Granted, I don't think that difference in probability is very high, and it's probably true that for most players, feel really is a decent heuristic for what's best for them. Perhaps my calibration on TheCore's suitability versus people's ability to reliably self-modify is too high. (I'm not putting down people's modification ability in general, I just find it unlikely that personal opinion/anecdote is more likely to be correct than TheCore-as-is, without evidence). That being said if one is a high level player and has a lot of in game experience, they may know ways to tweak TheCore for their style. Or, sometimes people are down right genius, after all, that's how most changes get implemented. For the record, I'm not saying that people should take TheCore as is like Dogma (I've critiqued it myself since the beginning), I think debate, critiscm and experimentation are part of what makes TheCore so practical. Im just of the belief that most changes that are good for one (baring things like changes based on ones hand) are likely good for most people too. Thanks for the reply
I think I have found a major difference in our thinking. Although it seems that we are both quite open-minded, you appear to base your ideas on quantitative evidence while I determine mine through intuition. To elaborate for you, my niche is in areas without the possibility of reliable proof.
Anyways, I'm glad we could reach a middle ground so easily.
On May 29 2013 08:33 Ninjury_J wrote:
Question: if you could layer control groups, (I.e. switch l and semi for layered cgs), what would you put on them? What modifications have you made?
I do not use layered control groups. It was pure theorycrafting. I might end up trying it, though.
I would probably end up placing tertiary army keys (lower in importance compared to basic attacking units and spellcasters) on them, such as Tanks/Widow Mines and Medivacs.
Oh, and also:
@Bulgogi: Thanks for initiating this enlightening conversation! :D
|
Correct! I tend not to trust my intuition, and don't trust others' generally what is your background, if I may ask?
|
I'm kind of all over the place.
My major is electrical engineering, which I am good at, but my current hobbies lie in psychology and storywriting. Besides Starcraft, lol.
|
Well, I'll throw in my 2 cents.
The Core is radically different from standard (and most other systems). If coming from standard, I'm skeptical that a person can empirically evaluate the core in a week or two. A person is probably looking at a couple of months and a couple hundred games of actually using the unmodified core as intended if they wish to do a fair comparison to a system they've potentially been using since the 1990's.
|
On May 30 2013 00:52 Smackzilla wrote:
Well, I'll throw in my 2 cents.
The Core is radically different from standard (and most other systems). If coming from standard, I'm skeptical that a person can empirically evaluate the core in a week or two. A person is probably looking at a couple of months and a couple hundred games of actually using the unmodified core as intended if they wish to do a fair comparison to a system they've potentially been using since the 1990's.
For a couple weeks of trials, of course you should be skeptical. That's why I largely disagree with the changes made by Bulgogi. The main reason I took him seriously is because he was not actually comparing TheCore to Standard.
|
On May 30 2013 01:22 Antylamon wrote:Show nested quote +On May 30 2013 00:52 Smackzilla wrote:
Well, I'll throw in my 2 cents.
The Core is radically different from standard (and most other systems). If coming from standard, I'm skeptical that a person can empirically evaluate the core in a week or two. A person is probably looking at a couple of months and a couple hundred games of actually using the unmodified core as intended if they wish to do a fair comparison to a system they've potentially been using since the 1990's. For a couple weeks of trials, of course you should be skeptical. That's why I largely disagree with the changes made by Bulgogi. The main reason I took him seriously is because he was not actually comparing TheCore to Standard.
Well, he doesn't really say what his point of reference is. Perhaps grid? Either way, I feel the same argument applies to grid or any system that lacks heavy reliance on pinky, thumb, layered cameras, and tabbed production.
|
I believe the merits of proposed modifications should be based on their actual intrinsic value rather than on judging the person who makes the suggestion. Who cares if the guy has been using The Core for a day or a year? This doesn't make the proposition any less (or more) interesting.
|
|
|
I guess I'm just taking issue with:
"I'll continue experimenting with alternative hotkey locations, but the current Core doesn't quite cut it for me. After making my changes though, I've played noticeably better."
Where issues or conclusions are based on personal experience, its hard to take those too seriously when that's based on 2 weeks of playing with a modified core. I'm definitely not making personal judgments. I just think its hasty to dismiss the vanilla core so quickly.
|
|
|
|
|
|
EPT
