"Da da da..."
"Da da da..."
On the screen, in that extremely simplified positioning test emulator, the crosshair constantly darted and weaved among several randomly spawning targets, eliminating them all.
The accuracy rate isn't actually that high.
but...
Sam, sitting in front of the screen, gripped the controller, controlling it almost entirely subconsciously.
The entire smoothness has been ingrained in my instinctive reactions; I can operate it perfectly based solely on intuition.
This is a pure controller-based shooting interaction.
"Amazing feel..."
After the round ended, Sam breathed a sigh of relief, looking at the screen showing an accuracy rate of over 80%.
For a player using a controller for the first time in an FPS game, this achievement is already a testament to exceptional talent.
However, Sam knew very well that there was probably a lot of exaggeration in this result. Because when he was operating it, he could feel a noticeable suction sensation, and some fine adjustments that he couldn't quite put his finger on but which played an important role.
It was as if he were on a shooting range, with a caring coach holding his hand and teaching him how to aim.
"How did you do that?"
Sam turned his head and looked at Lin Lixin, who was still checking the data output by the software.
This EM wasn't just for him to test the feel of the controls.
Similarly, Lin Lixin also needs to obtain more detailed data based on the output of the shooting practice software in order to further refine it.
This kind of thing can't be solved with cheat codes; it's entirely based on experience and can only be achieved by accumulating a large amount of data.
"Simply put, it's auto-aiming. Each enemy has a 'magnet' that attracts your crosshair, and with the algorithm's decision-making, you can quickly aim at multiple targets."
Lin Lixin recorded the data in a document while explaining it to him.
"However, the actual gameplay might not be as good as it is now, because the relationships between units are more complex, and it's not always possible to kill them with a single burst of fire. Player decisions will be more complicated. Therefore, in the next test, the health of each ball will be increased to 10."
The benefits of em testing are obvious.
Rapid iteration allows for quick implementation of any requirements and timely adjustments to projects. However, its drawbacks are also clear: it's too idealistic and too experimental.
When players actually start playing, things often get quite chaotic.
Just like that classic bar joke about test engineers.
It's impossible for designers to truly encompass all the ideas of every player; unexpected situations will always arise.
If players find this unexpected turn of events interesting, then it's a successful "feature," even making them think, "You even anticipated this?!"
Conversely, it will greatly ruin the game experience.
Just like your brother Ayu's Assassin's Creed Unity, which is worthy of being called a cult horror movie.
Test engineers need to understand that when a player comes to a bar, they might not order drinks. They might order fried rice, then jump on the Table to poop, and then make the bartender a whole plate of fried rice.
"Hiss... Speaking of which..."
Lin Lixin suddenly remembered something.
Who said Amenva doesn't have real-world player data?
" Cerny, what about the database used for information gathering on Quake 2?" Data is always the most valuable thing.
For Amenva, these things she had laid out long ago are now a huge experience package.
It could even be described as a vault.
Back in the Quake 2 era, Lin Lixin and Cerny had already tried to include a system that was barely noticeable in the game.
Information collection.
The game will automatically enable the capture function as long as the player selects the agreement that pops up upon first launch.
It records various states of the player in the game.
This includes detailed in-game data such as reload timing, hit rate, average elimination time, mobility, and the time consumed by a certain mechanism.
One or two pieces of this data may not be of much use.
But now Quake 2 is already the most popular FPS game among casual players.
A massive number of players contribute tons of data every day.
What's even more crucial is that this game significantly enhances the online multiplayer experience. Players naturally prefer to play online anyway.
This data will naturally be uploaded to the Amenva data center in accordance with the agreement requirements.
Previously, this thing wasn't very useful; it was just something we spent money to maintain as a database, in case we needed it someday.
And now it's that legendary "unexpected need".
Instead of having the testing team waste time and resources doing inefficient, repetitive testing, we can save a lot of time by using this readily available data.
Lin Lixin came to the workstation and quickly wrote a heatmap algorithm.
By processing these data samples into a heatmap, one can easily identify what he needs.
"Dead zone, acceleration, aiming assist strength, target adhesion..."
These things can be quickly applied to the current model.
However, they certainly can't just blindly believe everything.
These data contain many problems.
The most crucial point is that Quake 2 is a typical FPS game controlled by keyboard and mouse. Many keyboard and mouse settings differ from those for a gamepad, and this must be considered carefully.
Another difference is the sample group; Quake 2 has a certain learning curve.
This is especially true for the Arena mode.
Anyone who's still playing this game and is still completely absorbed in it is at least a top-tier FPS player.
The experience of these people may not be equally effective for newcomers.
Skilled players prefer higher sensitivity, smaller dead zones, and less camera slowdown, which allows them to react quickly enough.
However, for beginners, doing this will definitely make the camera spin so fast that it might even trigger severe vertigo.
All I can say is...
"The road to testing is long and arduous..."
Lin Lixin sighed helplessly.
On the bright side, at least now they can combine theory and practice.
" Cerny, I think we might need to set up a more professional behavioral lab in the future." "Hmm... With more and more online games becoming increasingly fast-paced, being able to build a professional team early on will definitely be beneficial for us."
Cerny nodded in approval.
Don't let his position as the team's architect fool you; he's a true full-stack developer.
Whether it's game development, hardware, or all sorts of other miscellaneous businesses, he's a whiz at it all.
A true frontline technician.
The two reached an agreement on this matter with almost no discussion.
"Network construction is also a problem... Once 'alf-life' is settled, maybe we can use that as an excuse to make a fighting game based on the new engine."
The name flashed through Lin Lixin's mind almost instantly.
The arena in Quake is excellent.
However, FPS games come in many forms.
Ultimately, the one who emerged victorious after market validation was indeed that very person.
"We need to catch up with web-based games too. Luckily, Activision and the others are nearby now, so I don't have to keep traveling back and forth to California anymore." "Speaking of which, I was thinking about this back when we were working on * Bloons Tower Defense *, whether that streaming data loading feature could..."
"Forest!"
Carmack's voice interrupted Lin Lixin and Cerny's conversation.
Lin Lixin glanced back.
"How?"
"During engine testing, we encountered an interesting little problem that I think you'll find fascinating."
"Oh ho?"
Lin Lixin raised an eyebrow, signaling Cerny to continue the tests for him and Sam, then turned and followed Carmack.
Along the way, Carmack gave him a brief overview of the situation in his own words.
"We've crammed a complete physics engine into the Source Engine, like... true physics, with rigid body physics, ragdoll effects, and all the other features you'd expect. So we used a room scene from Qake to test it."
"So what went wrong?" "We placed the NPC too close to the door. When the door opens, it hits the NPC's foot and gets stuck."
"What kind of question is this? Isn't this great? It shows that our new engine performs exceptionally well physically."
"No, no, no, that's the most amazing part. The physics for the door wasn't added this time; in fact, we tried to create rigid body physics for it back in the Qake era."
Lin Lixin was slightly taken aback.
"Perhaps the new module is too invasive, affecting the original calculations?"
"No, I checked. From a coupling perspective, this part of the code hasn't been touched at all. What's even more interesting is that I rolled back the code to the original version of the Quake 2 Engine, and the door still won't open."
Although he was explaining a rather serious matter, Carmack showed no negative emotions on his face; on the contrary, he was overjoyed.
Like a navigator who has discovered a new continent, he is excitedly reporting his achievements to the captain.
"Then the scene itself is problematic, isn't it? We tested it with the Quake 2 Engine and there were no issues."
"Don't rush, the most fun part is here. I ported the code back to NX. You know our architecture is interoperable, so porting it requires almost nothing. And then... guess what?"
Carmack was unusually excited today, even managing to keep everyone in suspense. "What?"
The door still won't open.
Lin Lixin was speechless for a moment.
Doesn't this further prove that there was a problem with the entire test room from the beginning?
"But!"
Before Lin Lixin could speak, Carmack spoke again, his tone shifting, a mysterious expression on his face.
"This door could be opened during the Qake 2 era."
"We've checked, and even using the original Qake2 components, this door still won't open properly. But during development, everything worked perfectly. I can guarantee that we have multiple archived internal test reports to corroborate this."
How could a newly copied file, with minor modifications, affect a completely untouched game source code backup stored on the server?
How is that possible?