679-08 | Yoh / PlaytestReviewsONSLACKER

From the professor:

I had fear that this was too ambitious of an idea for a project. And while it is impressive that you've gotten it to the point where you can see that a game can come out of it, I am still quite concerned that the pieces may not come together such that its a fun experience for the player.

1. Idea

I finally get the idea. However, its complex enough that it needs to be thought out and presented to the player in a way they can see what's going on in a meaningful manner.

I think with an idea this complex, the logical unfolding of the premise/mechanics/game is really an important piece.

2. Promise

OK, I can see why this could get to be fun. But it would have to work up to becoming increasingly challenging by leading up to it.

There is a lot of stuff going on. One of the potentially fun aspects is that you do a small set of things that cause lots of other things to happen. However, getting that feeling of connection might be a challenge.

3. Current State

There was a lot going on, but without lots of explanation I couldn't figure it out. Even with explanation, the connection between what I did and what happened seemed week. It was neat that all these guys were running around, but my connection to them (and the consequence of what they did) was a bit missing for me.

Many elements of the UI could use tuning as well.

4. Experience

For me, the real experience was finally learning what the game was about after weeks of not quite getting it. But actually playing felt light random clicking. And the abstractness lead me to not even remember where I had randomly clicked before. I felt like there was no reason to prefer randomly clicking one place rather than the other.

It was neat that there was a lot going on on screen, but I didn't feel connected to it. Guys might have been dying, but I could barely tell who were the good/bad guys, never mind develop a caring about not having my guys die.

5. Implementation

The implementation was clearly able to keep lots of things moving around on screen in a solid fashion.

However, the interface and gameplay issues really kept me from appreciating all that was going on behind the scenes.

6. Good/Bad

Good: I think the premise is really intriguing.

Bad: The premise is sufficiently complex that it will be extremely challenging to turn it into a good experience for the player.

7. What needs to be done

It appears that the good news is that the infrastructure is in place so that you can focus on the gameplay itself.

It will clearly take some thought as to how to better establish a connection between what the player does and what is happening. And to help the player build up an idea of how to make meaningful choices (without giving away too much).

Having the player work through a progressive set of levels is ONE stratgy for this.

I think that an improved UI that connects the two worlds, and makes the available choices and their consequences more obvious is critical.

From the TA:

Well, it's actually a game. I wasn't sure it would turn out that well, especially after the checkpoint, but it actually came together. :)

[promise] It could be a lot more fun, if you make some improvements.

[current state] It's cool, but it's also confusing. It takes a while to figure out what to do, and the feedback is really slow, since your army moves pretty slowly.

[game experience] It takes some fiddling around to figure out what you can do in the game. An explanation in the game would really help.

[implementation] It's pretty solid.

[good/bad] Good: It's a puzzle game. Yay! Bad: The pieces of the game don't fit together very well, and there isn't enough connection between cyberspace and the real world, and goal is unclear.

[if you had more time...] Establish strong connections between the two worlds. Label parts of both maps, to better see the relationships. Clearly mark your goal (at least so it doesn't look like just another room). Use the space you have better; the nodes are way too small to click on quickly. Improve the look of cyberspace... it's kind of plain and ugly... use color-coded circle nodes, maybe. :)

From your classmates:

After playing the game, I liked the idea of the game and its creative nature. Your goal is to get from point A to point B, and it’s the way you do it in which makes for an entertaining play.

This game would be really fun given a decent amount of time. This is because of the specialized content they must add for each node being hacked, coming up with interesting text and such. There would also have to be some degree of randomness for the hacking to add re-playability.

The game was still fun at playtest, but it needed some crucial changes which would make it really fun. There was some difficulty in coherency, and a general feeling of confusion, but this was do to some minor artifacts which could be modified quite easily.

There were no instructions yet, but the demo-er was good enough to explain what was going on. So yes I had to ask the instructor many questions, but I could see in the future the implementation of an on-screen tutorial, which is almost essential.

The implementation was pretty good. There were no crashes or anything like that. The only deal was I could not focus on my players and my actions at the same time. If I wanted to open a door, I had to look away from the action. When I completed an action, I had to click on it a few times because I had to look back at the game board to see what it actually did. Also the action text was really small, and should have been made larger somehow. The good part of the game was the text itself, and the effects of doing one thing or another.

What should be added is some kind of display which tells what part of the game world is what. This would make it easier to strategize and perform certain actions, because you could quickly associate the action with a location. If I want to open a door, close it, and then activate a turret, it should not be difficult to coordinate.

1. The overall idea of the game is fairly interesting. It gives you indirect control over mindless bots by changing the places they can go.

2. I'm not sure how good this game could be. If a completely immersive experience could be created, there would be a lot of promise. You need to be more connected with what is going in the "meat space".

3. The game is currently in a bad place. There is a definite connection gap between the two spaces. Also, it was very hard to tell what was going on. The hacking is too simplistic; a simple mouse click gets old. How the user interacts with the game is critical and right now the interaction is really lacking.

4. It wasn't very easy to get started. The controls were fairly simple, although they weren't documented. It was just tough to figure out what was happening, what you were trying to accomplish, and how you get to the goal. All that said, I was still determined to beat the level.

5. The game didn't crash and most things seemed to work. The only things overlooked involved the text size and the actual game play.

6. The game idea is pretty good and so is the simplicity of control. The game play itself and the connections shown to the user between spaces need work. It was just generally hard to figure out what to do.

7. Showing a connection between the two spaces is the biggest thing needed. Having the goal more clearly laid out would also be good. If there is time, the hacking should be made more interesting.

This is not the kind of game I like to play - I prefer the games that demand low latency responses and button mashing. Having "hacked" as part of an exercise in another course, there is a relationship at a high level to this game, but frankly, the exercise for the class was time consuming and difficulty. The fun part about hacking is knowing stuff you are not sopposed to know about the target system and watching as the administrator squirms. The hacking element of the game was about at the level of a script kiddie, with the simple "hack" button. I think that is the right level of detail for this game. In general, I don't consider hacking to be fun. Finding that buffer overflow isn't trivially and getting the jump-to-nop just right is frustrating.

The user interface was visually appealing, but the console output wasn't all in the right place (some was in the cmd window behind the game.) The font was tiny. It was difficult to tell the difference between good dudes and bad dudes. I had to zoom in to excess to find out. The level layout seems cool, and the AI is good enough, but the behavior is boring when there are not many dudes. Turrets were not easy to recognize and dudes appeared to die without reason, though further zooming revealing gunfire. I'd like to see some attention-grabbing on screen indication of where your dudes are dying. Consider the "your base is under attack" zooming rectangle indicator from starcraft.

The correlation between cyberspace and meatspace was important, but invisible, leading to frustration. There was a point where I stopped and asked what I was sopposed to do next, having not realized that there were nodes I had not yet hacked. Hacking itself was frustration, since I'd click repeatedly to hack, then not stop immediatly upon success and accidently open some door or enable some turret. Selecting a "door" in cyberspace gave no indication of its state in meatspace, and the first input option ("open" or "close") was always initially selecting, implying, erroneously, that the door was in that state.

There should have been more levels. This game mandates training levels and documentation.

I think the game crashed once, but I don't think it was during gameplay.

Onslaker struck me as a game with great potential, but still needing a lot of cleaning up to make the gameplay experience fun. The major issue is communicating to the player what he needs to do. I was able to pick up the basic idea fairly quickly (although the goal room should be better marked), but still had to fight against the UI in order to accomplish what I wanted. Planning the strategy of how to best get your units safely to the goal room is neat, and provides the player with many possibilities. However, even once I knew what I wanted to do ("i should really open that door - and then turn off the turret") it was frustrating to do it. If I've successfully broken into a node and can control some element of the map, it should be clear to me and easy to get back to. Some way of signaling the relationship between the network and real world maps is a must. The actual process of 'hacking' feels very arbitrary, and could very quickly turn off an impatient player. When the whole thing is random I have no feeling of whether I should sit and click mindlessly with hopes of getting in, or if I have no shot. At the least, it should be indicated to the player what the current level of security is on a system so they have an idea of their chances. I think it would be really cool if there were some actual gameplay involved with breaking into a system - even the little puzzle with the toaster was more fun than clicking hack over and over - but I understand we are on time limitations. Finally, some sense of urgency could make the game more fun as it presents the player with a clear challenge. Perhaps having a countdown until a 'bomb' goes off, or the player is tracked, or something of that nature would get the players nerves going and provide a more intense experience. Balancing a timer like that will be very delicate though. Over all, I think this game could be awesome. I didn't quite know what to expect going in, but was very pleasantly surp rised. It just needs to be more accessible to avoid frustrating players (and DON'T keep in the node that resets the entire network...not funny)

1. The game idea is a potentially interesting one; it holds promise. At the same time, I think that making the game fun would be difficult. It is extremely dependent on good level design.

2. Again, this game is dependent on levels. I don't know how your levels are represented and if you can easily create new levels.

3. The current game is okay, but not great. I found that most of the time I was pretty much randomly clicking, randomly finding the next node. It would probably have been possible to take notes and actually figure it out, but I had neither paper nor the inclination at the time to do so. (I almost wonder if there's a way to *make* the player pay better attention; this might improve things as the player stops randomly clicking, but is also probabyly risky.)

4. I don't recall asking questions, but there was a fair amount of unsolicited advice given. This ranged from the general idea of the game to the specifics of the "victory switch" to which nodes should be explored next or couldn't yet be hacked.

5. The main problem I have with the current implementation is that I think it's difficult to click on the nodes in the graph. There isn't any leeway around them to miss, and the nodes are very small; it would often take a couple attempts to hit them.

The other usability issue is that if you selected a node, did something with it (for instance disabled a turret), selected a different node, then went back to the first, your initial selection wasn't displayed. The state was the same as it was, but the visual indication wasn't up-to-date or correct.

I did eventually get it to crash, but (1) this was after I had won the game and (2) I had done something unusual. (Unfortunately, I forget exactly what.) This didn't interfere at all.

6. I think the look and feel of the game fits very well with the premise. I liked the subwindows a lot (though if things were resizable it would have been even better), and the look of the widgets. Puzzle games are often good, and I think this has a fair bit of that flavor.

7. I think that one of the most important things is an indication on the map of what the currently-selected node controls (when applicable). For instance, if it opens or closes a door, that door should be indicated on the map. (The node that controls an unspecified door shouldn'd do this until you flip it a couple times, and maybe change to another node, but then it should. By that point the player should know what it controls.) Even when it said "this controls the door in sector whatever", since there isn't a visual indication of where sector whatever is, I just would flip the door's state to find it.

Even better would be if you could select something (door or turret) on the map and have it select the node in the network. This would have helped with a lot of the randomly selecting nodes I did in search of the one that controlled the door I wanted.

As it stands now, I almost think I would want pencil and paper to play again so I could keep notes on what node controls what.

For nodes that don't have direct correspondences, it would be nice to have in indication in the network as to the function. For instance, have the node say "firewall control" or something.

Idea: Seems a bit odd, it is a unique idea, but DotA is fun so it could be good.

Promise: If the effects the player has on the world are interesting enough, and it's interesting to cause those effects, it could be fun.

Current State: Short, only one level, and the lack of feedback basically makes it a trial and error game.

Experience: No in-game documentation, which really hurts a complicated game like this. I didn't really know what was going on without a lot of nudging from the demoer.

Implementation: Seemed solid, no crashes. Was no indication other than a console message that the level had been won.

Good Parts: Strategizing was fairly fun, figuring out which doors to open/close to direct my guys towards the end.

Bad Parts: Complete lack of feedback- I needed to spam click every node without any knowledge of whether it is hackable or not. I also needed to click open/close a few times just to see which door the node controls. No indication of what each node does without clicking it, and the menu defaults to the top every time so I don't even know the current state of the node.

Things to be added: More interesting attack indications. More levels. More abilities to affect the world.

1. Idea: Once it was explained to me what is going on, I liked the idea. Controlling the battle indirectly is a good idea, but implementing it and keeping the player's attention is hard.

2. Promise: I think this game could be very good with enough time.

3. Current State: It's alright to play right now. It was a bit frustrating at first cause I didn't know what I could do or how it was affecting the game.

4. Experience: Like I said above, I didn't know what I was doing for a while. I'm glad there was somebody giving the demo, cause I would have gotten to frustrated to continue.

5. Implementation: No crashes occurred and everything ran smoothly on screen.

6&7. Good/Bad&Fix/Add: I really like the idea of indirectly controlling a space battle. I think the interface should be made more clear. The text should be bigger and the nodes that control doors, turrets, etc... should be color coded with the battle onscreen so the player clearly know what node is connected to what door/turret/etc. Work should also be done in the area of teaching the player how to play. This might be done over the course of many levels.