Cockatrice - intended as successor to MWS

[JDawg13]

Fair enough, thanks for the response.  Even with the small cards this is still an awesome program.

[Gekoratel]

My friend was trying to install the application on his Mac with my walking him through it and he got the same error message as Tamriel on the Cockatrice forums.  I know that he installed QT 4.5 because he was able to open up the QT GUI but it was a bit confusing about what to do with it.  Through the command line when he tried man qmake and man make it said "no command found".  I was wondering if any progress has been made with getting through this issue on macs or if there has been a pre-compiled version ready for the Macs.  Looking forward to playing against my friend, thanks.

[AmbivalentDuck]

You need the "xcode" package from apple.  Ie. Libraries are useless without a compiler.

[Gekoratel]

I downloaded the xcode package from apple but am still having issues trying to get this working on a mac, I followed the step by step guide on the forums but it fails at the qmake step.  I'm trying to walk a friend through without seeing the computer so its not the easiest.  Here is my post from the forums, which details the steps I went through.

Thanks for the post Ztyxx, sadly my friend is still having some difficulty getting it working on his machine.  We installed GIT, QT, and XCode; all of which seem to be installed properly.  After that we are able to GIT the file onto his machine but then through Terminal we get the following error.

On the following line: "cd cockatrice/cockatrice; lrelease cockatrice.pro; qmake; make -j2" after updating the header file the lrelease seems to work properly generating the following output.

Updating '/Users/XXX/cockatrice/cockatrice/translations/cockatrice_de.qm'...
    Generated 373 translation(s) (373 finished and 0 unfinished)
Updating '/Users/XXX/cockatrice/cockatrice/translations/cockatrice_en.qm'...
    Generated 1 translation(s) (1 finished and 0 unfinished)

When we type in qmake; nothing happens at all just a blank return.  Then when we type make -j2 it generates the following error message.

make: *** No targets specified and no makefile found.  Stop.

Thanks for the help anyone.

[AmbivalentDuck]

That actually sounds about right.  You have to make sure you're in the directory with the makefile. 

Maybe try "make" instead of "make -j2"

[Gekoratel]

Thanks AmbilvalentDuck, I'll see if my friend can try it again today and I'll post here about how it works.

[mbruker]

Your problem is that qmake doesn't actually generate a makefile when invoked without parameters on MacOS. You have to write
qmake -spec macx-g++
to make it work.

[Gekoratel]

Thanks mbruker that got us a step closer, I'm cross posting this from the cockatrice forums

Thanks Brukie I'm pretty sure that worked since it went through a lot of steps building the application. Once its finished running what are we looking to run exactly. My friend looked around for any .exe file to run but didn't see anything. When he went to open Cockatrice.pro it opened in Qt probably because its a project file or something.

Also in the oracle folder we didn't see a contents folder in order to move the necessary binary file unless I am not understanding that step. But either way we need to figure out how to open the application itself. Thanks the help guys.

[CorwinB]

A small heads up : you can already import ROE cards into Cockatrice by putting the following section in your sets.xml file :

Code:

        <set import="1">
            <name>ROE</name>
            <longname>Rise of the Eldrazi</longname>
        </set>

and then relaunch Oracle.exe

Regards,

[AmbivalentDuck]

Ubuntu build presently fails:

src/carddatabase.cpp:348: error: ‘class QNetworkRequest’ has no member named ‘originatingObject’

"qmake -v" produces

QMake version 2.01a
Using Qt version 4.5.2 in /usr/lib

I'm guessing that Qt 4.5 is the problem?

[BruiZar]

I looked at the screenshot but I'm not really impressed with the way it looks. For me the only reason to use a program such as mws (which i quit) is if the interface works and looks nice, everything is intuitive, and i dont get headaches from the tap / play card sounds. Before I try this, this needs to get polished up.

[mbruker]

Ubuntu build presently fails
I'm guessing that Qt 4.5 is the problem?

Sorry, I accidentally broke that. Qt 4.5 really is the problem here, but while 4.5 has some inherent problems that prevent Cockatrice from working perfectly anyway, I don't intend to stop supporting it altogether. I hope I can fix it this evening.

[mbruker]

Compiling with Qt 4.5 hopefully works again. Please test.

[UniversalSnip]

For me the major concerns in a free magic program are the interface and stability, in that order.

MWS: Good interface, awful stability
OCTGN: Awful interface, good stability
This: Awful interface, ??? stability

Until I get a program I consider usable I'm just gonna have to stick with mws.

[mbruker]

Would you be so kind as to explain what makes our interface awful, in your opinion? You see, we can't fix what we don't know is broken.
Again, I'm sorry for the currently very slow development. I hope there will be more time soon.

[TheShop]

I am a blank slate as far as computer skills are concerned, but I really want to use this program.  I currently use lackeyccg(okay for goldfishing, but then again I can goldish on microsoft excel...problem is, no one plays vintage on their server). I have a mac, and as exciting as this is, I think I am going to have to wait until someone figures out a way to make this much more user friendly to install.  I know I can bootcamp this and run native on PC, but I need to do some updates to do that...and is it really worth partitioning your hard drive to run a single program in native pc?

Any news on the Mac compiling front?  Also, I clicked your download page and could not get a download to work, but that was around the time people were having issues with 4.5 or whatever.

[AmbivalentDuck]

Now works on Ubuntu 9.10.

mbruker, can you write a ./install script?  It can't be that hard to write a script that handles executing lrelease, qmake, and make for oracle and cockatrice.

Important OMITTED defaults:
create and automatically set pics directory: ./pics
create and automatically set decks directory: ./decks
create and automatically set cards: ../oracle/cards.xml

As far as mac, once ANYONE has built it, it should be trivial to make a dmg for it.  Sadly, Macs are pretty crappy about dependency management.  They're built for the same closed-source development model as windows.  They just avoid dll hell.

[mbruker]

The install script should be easy to make. I guess I should switch to an autoconf/automake kind of compile process.
Making a MacOS release is the first thing I'm going to do as soon as there is some free time. I managed to get a running Mac system, so that shouldn't be much of a problem. However, it's not as trivial as you're saying because a) one has to make a universal binary so that it will run on different hardware, and b) some changes might have to be made in the code to simplify installation (that applies to all OSs, of course). Obviously, parts of the code are still in a proof-of-concept stage, so that needs to be changed.

[UniversalSnip]

Would you be so kind as to explain what makes our interface awful, in your opinion? You see, we can't fix what we don't know is broken.
Again, I'm sorry for the currently very slow development. I hope there will be more time soon.

With the caveat that I'm judging just on screenshots, I'd say it's because it's because you don't have somebody really interested in interfaces in charge of them . But that's the root problem. Here's a specific example:

http://img.photobucket.com/albums/v410/UniversalSnip/cockatriceplayscreen.jpg

everything I boxed out is wasted space for one reason or another.

[mbruker]

Thanks for your reply.
Concerning the two biggest boxes, I can see why you think they are a waste of space. This space is not really wasted because it is used up as the table grows, but I hope we can find a nicer looking solution for that as I've heard that argument a lot now.
Why don't you like the card information box on the right? You have to keep in mind that the oracle text of a card is not necessarily equal to the one printed on the card, so the picture alone is not enough.
The rest of the boxes I can get behind, but I don't really know what to fill that space with. Any suggestions are welcome.

[Delha]

FWIW, I consider dead space to be a minor issue (and would call it a layout issue instead).

When I heard the word interface, I there were flaws which impeded effective communication between the player and the program. Stuff like excessive confirmation dialog, or similar issues. Basically, anything that would require the player to put unnecessary effort into a given action.

I haven't tried Cockatrice and therefore won't comment on it in specific. As a general rule though, I care more about ease of use. If the layout problem were something like say, putting the library/graveyard on the opposite side of the chat window, I would feel it was a major concern. A few square cm of blank corner space, not so much.

Addendum: Have you checked out BrettSpielWelt? I've only played Dominion over there, but it might be a place to find ideas.

[UniversalSnip]

Thanks for your reply.
Concerning the two biggest boxes, I can see why you think they are a waste of space. This space is not really wasted because it is used up as the table grows, but I hope we can find a nicer looking solution for that as I've heard that argument a lot now.
Why don't you like the card information box on the right? You have to keep in mind that the oracle text of a card is not necessarily equal to the one printed on the card, so the picture alone is not enough.
The rest of the boxes I can get behind, but I don't really know what to fill that space with. Any suggestions are welcome.

Well, my main priority is not looks, but these kind of things quickly add up to create usability issues. That card box is actually the least of my issues (for example, where are the binds? These speed up play considerably by letting you communicate without taking your hand off the mouse. Why are the mana counters disproportionately favored? Why are the extremely useful step buttons not given a central position? etc etc)

I did consider the oracle text issue before boxing that out and thought of a few ways you could handle it. Here's my favorite solution:


Card Display Panel

Description
This displays the image of the last card moused over. It is surrounded by a gray bevel and has two modes:

- Render mode. This displays the card using a combination of the card skin, oracle text and the partial scan image file associated with the card, if any.
- Scan mode. This displays the image file associated with the card, if any. For cards with an image file, this is the default.

Mockups
Default:
http://img.photobucket.com/albums/v410/UniversalSnip/CardDisplayPanelScanMode1.jpg
Moused over:
http://img.photobucket.com/albums/v410/UniversalSnip/CardDisplayPanelScanModeMousedOver.jpg

Behaviour
Mousing over the panel turns the beveling red, indicating this element is clickable. Clicking it toggles the display between render mode and scan mode.

Tooltips
"Click for oracle text" when in scan mode.
"Click for card scan" when in render mode.

FWIW, I consider dead space to be a minor issue (and would call it a layout issue instead).

When I heard the word interface, I there were flaws which impeded effective communication between the player and the program. Stuff like excessive confirmation dialog, or similar issues. Basically, anything that would require the player to put unnecessary effort into a given action.

I haven't tried Cockatrice and therefore won't comment on it in specific. As a general rule though, I care more about ease of use. If the layout problem were something like say, putting the library/graveyard on the opposite side of the chat window, I would feel it was a major concern. A few square cm of blank corner space, not so much.

These are not separate issues.

[mbruker]

where are the binds? These speed up play considerably by letting you communicate without taking your hand off the mouse

Do you mean text messages bound to keys/key combinations? They can be configured (though I have to admit that this needs some improvement).

Why are the mana counters disproportionately favored?

So that you can read the numbers well.
You have to understand that horizontal space is not an issue as the size of the table is usually determined only by the height of the window, barring a very unusual aspect ratio or a lot of cards on the table (making it grow horizontally). So we don't need to feel bad when we "waste" space with big mana counters as long as that space isn't above or below the table.

Why are the extremely useful step buttons not given a central position?

As explained in the last paragraph, that would be a waste of space. However, since I've had this discussion before multiple times, I will make it configurable so that you can all have your gigantic black space on both sides of the table and your unrecognizable, fingernail-sized cards like they have them in MWS.
Apart from that, I would really like intuitive key bindings for the phase buttons. I recognize that moving the mouse pointer there every time you want to change phases is too much of an effort. I think I might bind them to the F-keys (obviously changing the bindings for text messages), or Alt+number, or something like that.

- Render mode. This displays the card using a combination of the card skin, oracle text and the partial scan image file associated with the card, if any.
- Scan mode. This displays the image file associated with the card, if any. For cards with an image file, this is the default.

While I appreciate your thinking about the topic and explaining your ideas, I don't quite understand what the benefit of this particular idea would be. Wouldn't the interface have the same features that it has now, just with more mouse clicks necessary? I quite like hovering over a card and immediately seeing its attributes without having to change the mode of the display zone first... You see, there is again no space wasted as you wouldn't have anything else to fill it with, would you? Please correct me if I misunderstood what you were saying.

[UniversalSnip]

where are the binds? These speed up play considerably by letting you communicate without taking your hand off the mouse

Do you mean text messages bound to keys/key combinations? They can be configured (though I have to admit that this needs some improvement).

Nope, I mean buttons (although keyboard shortcuts are great as long as they're not made necessary). Anecdotally, the most used elements of the game window are:

1) The cards in hand/on the board
3) Step buttons
2) The 'end my turn' bind
3) Portrait, for counting life up and down
4) Opening your library

Why are the mana counters disproportionately favored?

So that you can read the numbers well.
You have to understand that horizontal space is not an issue as the size of the table is usually determined only by the height of the window, barring a very unusual aspect ratio or a lot of cards on the table (making it grow horizontally). So we don't need to feel bad when we "waste" space with big mana counters as long as that space isn't above or below the table.

I'm not sure I really understand this. When I get home in a few days I'll try the program out and have some more feedback there... the philosophy seems suspect there. There is always something useful you can do with excess, even if it's just framing important elements in white space.

On a side note, if there's an adequate snap to grid feature, in my experience almost everyone prefers to align their cards horizontally.

Why are the extremely useful step buttons not given a central position?

As explained in the last paragraph, that would be a waste of space. However, since I've had this discussion before multiple times, I will make it configurable so that you can all have your gigantic black space on both sides of the table and your unrecognizable, fingernail-sized cards like they have them in MWS.
Apart from that, I would really like intuitive key bindings for the phase buttons. I recognize that moving the mouse pointer there every time you want to change phases is too much of an effort. I think I might bind them to the F-keys (obviously changing the bindings for text messages), or Alt+number, or something like that.

Binding text is a great idea, but you should already be using the F keys for game functions like start new game. Having the steps in the middle is also important because they're shortcuts to common actions, ie drawing with the draw step button or untapping with the untap step button.

- Render mode. This displays the card using a combination of the card skin, oracle text and the partial scan image file associated with the card, if any.
- Scan mode. This displays the image file associated with the card, if any. For cards with an image file, this is the default.

While I appreciate your thinking about the topic and explaining your ideas, I don't quite understand what the benefit of this particular idea would be. Wouldn't the interface have the same features that it has now, just with more mouse clicks necessary? I quite like hovering over a card and immediately seeing its attributes without having to change the mode of the display zone first... You see, there is again no space wasted as you wouldn't have anything else to fill it with, would you? Please correct me if I misunderstood what you were saying.

It eats up space in the history log. Scrolling up is used all the time... or it would be if it weren't so torturous (I suggest a way of opening the history in a text file). Ironically I'm a lot more likely to be looking through the history if I have a rules argument than I am to need the oracle text, especially since recent cards are disproportionately represented in games.

[mbruker]

3) Step buttons
2) The 'end my turn' bind

These two are actually supposed to be the same. It would be nice if you could press some F key to move to the end-of-turn step, indicating to your opponent that you're done.

I'm not sure I really understand this.

On a side note, if there's an adequate snap to grid feature, in my experience almost everyone prefers to align their cards horizontally.

Let me try to explain. The table is divided into four rows with a theoretically unlimited width. Although you can use them for pretty much anything, their intended purpose is:
1) lands
2) non-land, non-creature permanents
3) spells
4) creatures.
Whenever you have the game place a card from your hand automatically on the table, it will choose the leftmost free spot in the respective row (the screenshot you referred to was slightly outdated). This fills up the table at some point, so it has to grow horizontally. However, it never grows or shrinks vertically (it could be allowed to shrink if I were to impose my playstyle on the users, but maybe they want to use two rows for lands or whatever). Because of that, the zoom factor of the table (=> recognizability of cards) depends almost solely on the vertical size of the area where it is displayed. Usually, it depends neither on the number of cards on the table nor on whatever space is wasted on the left or right side of the screen.
You might notice that I borrow a lot of ideas from MTGO, though some of them are not sensible without the program knowing anything about the game. I find their solutions to these problems to be very good, though.

Binding text is a great idea, but you should already be using the F keys for game functions like start new game.

F2 -> concede
F3 -> view library
F4 -> view graveyard
F5 and up -> text messages
I think we need to change these so that all phases are accessible via F keys. As soon as 'end my turn', 'move to attack' and so on is done via change of phases, we don't need so many text messages any longer, so they can be bound to something like Alt+number or Ctrl+number.

It eats up space in the history log. Scrolling up is used all the time... or it would be if it weren't so torturous (I suggest a way of opening the history in a text file). Ironically I'm a lot more likely to be looking through the history if I have a rules argument than I am to need the oracle text, especially since recent cards are disproportionately represented in games.

You're right about this: I'm just not used to playing against strangers on MWS. I've almost never had a rules argument in a Cockatrice game.
I think it won't hurt making the behaviour of the card display zone configurable so you could hide the oracle stuff if you know it anyway. We could also do something like the middle-click feature in MTGO (it displays an enlarged version of the card where your mouse cursor is). Then you could be even given the option to hide the card display altogether, since what do you need the picture for?

By the way, I'm sorry that this is quite a theoretical discussion. Although thinking about these topics is very valuable, there's really not much else I can do at the moment because of the lack of spare time

[UniversalSnip]

3) Step buttons
2) The 'end my turn' bind

These two are actually supposed to be the same. It would be nice if you could press some F key to move to the end-of-turn step, indicating to your opponent that you're done.

So when you click the end step button it says 'end my turn' in the text box? I like that

I'm not sure I really understand this.

On a side note, if there's an adequate snap to grid feature, in my experience almost everyone prefers to align their cards horizontally.

Let me try to explain. The table is divided into four rows with a theoretically unlimited width. Although you can use them for pretty much anything, their intended purpose is:
1) lands
2) non-land, non-creature permanents
3) spells
4) creatures.
Whenever you have the game place a card from your hand automatically on the table, it will choose the leftmost free spot in the respective row (the screenshot you referred to was slightly outdated). This fills up the table at some point, so it has to grow horizontally. However, it never grows or shrinks vertically (it could be allowed to shrink if I were to impose my playstyle on the users, but maybe they want to use two rows for lands or whatever). Because of that, the zoom factor of the table (=> recognizability of cards) depends almost solely on the vertical size of the area where it is displayed. Usually, it depends neither on the number of cards on the table nor on whatever space is wasted on the left or right side of the screen.
You might notice that I borrow a lot of ideas from MTGO, though some of them are not sensible without the program knowing anything about the game. I find their solutions to these problems to be very good, though.

I'll just have to try it.

Binding text is a great idea, but you should already be using the F keys for game functions like start new game.

F2 -> concede
F3 -> view library
F4 -> view graveyard
F5 and up -> text messages
I think we need to change these so that all phases are accessible via F keys. As soon as 'end my turn', 'move to attack' and so on is done via change of phases, we don't need so many text messages any longer, so they can be bound to something like Alt+number or Ctrl+number.

There are 13 steps so that could be a bit awkward. What would probably be better is to bind maybe the four most useful phases to shift + number keys and make tab the 'next phase' button.

It eats up space in the history log. Scrolling up is used all the time... or it would be if it weren't so torturous (I suggest a way of opening the history in a text file). Ironically I'm a lot more likely to be looking through the history if I have a rules argument than I am to need the oracle text, especially since recent cards are disproportionately represented in games.

You're right about this: I'm just not used to playing against strangers on MWS. I've almost never had a rules argument in a Cockatrice game.
I think it won't hurt making the behaviour of the card display zone configurable so you could hide the oracle stuff if you know it anyway. We could also do something like the middle-click feature in MTGO (it displays an enlarged version of the card where your mouse cursor is). Then you could be even given the option to hide the card display altogether, since what do you need the picture for?

I like the idea of just eliminating the card display panel and reducing it to pure mouse overs. Then the history could fill the entire right side of the screen except a long button at the top labeled 'oracle text' which would expand downwards when clicked. Very professional!

By the way, I'm sorry that this is quite a theoretical discussion. Although thinking about these topics is very valuable, there's really not much else I can do at the moment because of the lack of spare time

My prayers would come true if another magic program got a really smooth, sharp interface... MWS's instability and shareware nagging drive me nuts.

[mbruker]

I finally managed to get a Mac system and made a Cockatrice package for Mac.
There haven't been many changes to the git code. I'd just like to have this tested so that I know what to change for the next release with regards to packaging and platform specific things, so I'd appreciate it if someone gave this version a try.

Here's how it should work:
1. Download this file: http://cockatrice.de/files/cockatrice_mac_20100510.dmg.zip
2. Unzip, mount, copy the folder that is inside somewhere on your disk, and unmount.
3. Start the oracle application inside the folder. In the 'File' menu, select 'Download sets information' and click OK. Once the list of sets has been retrieved, click 'Start download' and wait until it is finished.
4. Start Cockatrice and play. Adjust settings if necessary. The paths you absolutely need to set for it to work should be auto-detected though.

There will be a nice little installer for Windows also.

[Mindstab_Thrull]

Warning: I have not yet downloaded Cockatrice as of this post.

My big question regarding Cockatrice is this: How hard is it to use decklists posted in other formats - .dec (Apprentice), .mwDeck (Magic Workstation), or .txt (from Magic Online)? I'd really rather not have to download another program and then find I have to enter all the decks I have manually. If it supports Apprentice and/or MWS formats it helps boost its usability immensely, even if it has its own native format.

Mindstab_Thrull
Im in ur brainz, nommin ur sanity

[CorwinB]

Warning: I have not yet downloaded Cockatrice as of this post.

My big question regarding Cockatrice is this: How hard is it to use decklists posted in other formats - .dec (Apprentice), .mwDeck (Magic Workstation), or .txt (from Magic Online)? I'd really rather not have to download another program and then find I have to enter all the decks I have manually. If it supports Apprentice and/or MWS formats it helps boost its usability immensely, even if it has its own native format.

Mindstab_Thrull
Im in ur brainz, nommin ur sanity

Cockatrice supports .mwDeck files.

[TheShop]

It is up and working on my computer but i am usually alone on the server.  When I get back from vacation I may try to recruit or coordinate with others.

Props- simplicity and speed here are better than my previous mac compatible programs

Slops- it did take a minute to figure out (I'm not a computer genius) how to use this.