Best rts is league, simply a fact, it outperforms sc2 by a factor of 1400x. David Kim did a pretty decent job by contrast to the balance counsel who are probably huffing paint fumes as we speak. He screwed up sc2 at the end because he saw korea going back to broodwar and so he tried to make SC2 mimic broodwar & this was catastrophic because the game market was crystal clear that that style of RTS is dead as a doornail. Korea lost interest in SC2 because the pro scene collapsed as a result of the region lock.
Because of these catastrophic mistakes, SC2 and Broodwar began competing for the same niche group which was already small to begin with. That means cutting the talent and funding right down the middle. Thatâs how SC2 died. He realized his mistakes because heâs designing Battle Aces with those considerations in mind. But itâs too late. Gamers are like coke addicts. They get 1 taste of a game and never stop playing it and thatâs why breaking a new game into the market is incredibly hard. I give him 9 in 10 odds of failure because heâs trying to compete against league for the casual rts players. Those players will probably play league until they die of heart failure at the ripe old age of 127.
But no, I wouldnât want to be the guy. He will probably turn a profit. He might be able to buy a house with that profit. But itâs not going to be what SC2 was in that it dominated the RTS market with 1 million+ active players. I wouldnât want to design sc2 either. The game has gone so far off course that it canât be saved. Itâs doomed. I knew this would happen literally 10 years ago at this point and, while David Kim has figured it out, the people who now manage SC2 still donât get it. They are still designing the game around the mechanical skills of Serral and Dark. The only value SC2 has at this point is to be a comedic reminder of exactly what not to do if you want to be successful. At this point, if I were onboarded to save SC2 I would refuse because I wouldnât want my name associated with it & the people running it. It would legit delete my reputation. Nobody would ever hire me ever again. So no, I donât want to be David Kim & save SC2. I could absolutely do it. I am probably one of the few people on the planet who could. But never in a million years am I going to associate with people who prioritize the pro scene of the player base. It makes you radioactive. Nobody in their right mind would ever trust your judgement ever again.
I loved watching the casters at Homestory cup eat crow. One of the most satisfying moments of my life. Toss throws the unthrowable of unthrowable series, and they are thinking in their minds âyeah this looks baaaaaadâ because it does. Toss has been buffed to the point someone can get 7k mmr with 4 gas mass zealot that he just amoves at the terran and thereâs literally nothing the terran can do to stop it except to pray the protoss plays horribly. Iâve said it for years. They called me names on their various alt accounts. Lowko made videos making fun of me and insulting me. Well shucks looks like I was right all along. Eat crow, losers. I wonât save SC2. You could offer me a million bucks and Iâd give you the middle finger. Itâs doomed and it is 100% the balance counsels fault. You want me to save it? That ship sailed. Lmao. In no universe would I want to be David Kim.
Thatâs not a threshold, thatâs a reason (and not a very good one, Terran has excellent defenses, but thatâs the issue, you canât defend as Terran, Terran is all about momentum).
Thatâs a lie, lol.
Right. Terran players are casual because of the lower MMR. It proves itself!
Funny coming from someone who probably pulls 45k on a good year. Give me a break.
Then please explain why its Not a good one. A decent sc2 Player knows that He can counter every single unit in the Game with Marines. That is a fact. But in Order to do it you need to micro it very very very well.
You say terran is a momentum based Race. Meaning those 2 Statements (momentum based and Marine Not being a good reason for t being hard) pretty much contradict each other.
I actually have agreed to you that you need to be aggressive and use momentum. What is the Unit that gives us the momentum and Aggression? Exactly. Marine in Combination with medivac. This is exactly what i Said.
Hahahahhahahahahaha
Sim City is a new word for you then terran has a much much Stronger defense than toss. Or do you mean Defensive Style ? Then Turtle mech is a new Style to you ? Obviously viable to at least 6.5k mmr against every Race.
Please use Arguments. That is Not a lie. Its a fact. Bronze and silver is flooded with terrans. Those are Casual Players who belong there. Thus making the Statement: terran gets more often picked by Casual Players is indeed correct.
Im Sure you will understand it as: terran Players are more Casual. But this is Not the Case.
And because of their Activity apm and spm.
Oh my oh my. You are so defeated that you need another Personal Attack. Adressing my sexuality and also my income. Im so offended now.
Marines suck without micro, theyâre absolutely awful. In the hands of a Master? Theyâre the best core unit in the game.
Try to turtle mech vs Protoss lol. Itâs way harder than just doing a bio tank push. Besides now youâre contradicting yourself. Defending without Marines is incredibly difficult (Marine/bunker is probably the best defense in the game, aside from maybe Queens), but if you turtle with Marines, then you give up momentum which you need as bio.
My point was that turtling as Terran is terri-bad. You have to attack and use momentum.
Terran has more activity, more APM, and more SPM on averageâŚ
Yeah this is the first time that weâve traded blows. Stop playing the victim, weâre here because of you.
THIS IS EXACTLY WHAT I SAID.
In short: the higher you are the better you have to be with marine. This was exactly my point. jesus. you have to be very good to abuse it to the max.
The marine is the best unit hands down. BUT ONLY IF YOU MICRO. WTF. DUDE. Im losing my mind right now
Did it. Its easy. You can also look at my nephest. Took me 1,5 month from being a complete noob with terran to about 4.1k. It was with mech.
It easily works up to 6.5k. I gave a friend of mine the advice to just use turtle mech against zerg. his mmr went up by 500 mmr. If you dont have the mechanics you better use mech terran.
Compared to toss but less compared to zerg. But the point was withing their own race compared to bronze or gold, the silver player is between them. Meaning there are measuarble metrics that a silver terran is indeed worse than the gold terran is not held back by anything other than their own activity. spm or apm.
I want you to grow. You are clearly in the wrong. You have 0 convincing arguments. You only have âBut on average terran has less mmrâ and every decent person can figure out the massive flaws of this statement. You even change your argumentation to: Some pros say terran is hardest so pleeeeeeaaaassssssseeeeee agree that terran is somewhere somewhat harder but how and why and where we dont know but just please please please agree
Just checked. All of SC2 channels on twitch have a net sum of less than 400 viewers right now and a single channel for Battle Aces has over 600. The final day at HSC 25 doesnât make their top 10 video list. Itâs #16. Past broadcasts â sort by popular â count down to 16. What happens if we do the same test on GSL vods? Their latest video ranks #177th in number of views. Ouch. The balance counsel are just so proud to show their new balance patch & how fun it is for protoss to be strong again! And then reality hits them like a freight train.
The balance counsel gave the middle finger to their customers & now they are surprised the customers left. How about that. Then someone asks âHey, you wanna be David Kim & save starcraft 2!â. Been there, done that. The only thing saving sc2 does is get you forum reports from forum karens trying to turn the internet into their personal safe space. Oh and -1200 karma on reddit. Oh and multiple people harassing you & calling you names, lying about you 24/7, etc. The SC2 community is sick at the very core. Itâs like a dying animal that bites anyone who tries to help it. Thatâs a good analogy.
Iâve coded entire renderers & graphics engines from scratch and some crazy just says I am wrong, and tells people to report me for being a troll. Itâs like a monty python skit. I couldnât believe my eyes. How is it even possible to be that wrong. Itâs pure unadulterated hatred. Thatâs what it is. Itâs the only thing that can make someone that blind.
You know, out of all the comedians out there, and there are a lot of good ones, the one I identify with the most is Bill Burr. He will be cracking jokes, laughing, and then he will get serious and go on this long a** rant about how much he hates everything. Thatâs a great guy. Iâd love to buy him a beer sometime. Thatâs a guy to go drinking with.
I have a feeling that if the pro scene was really dire the balance council will take more drastic measures to shake up the meta. Iâm fairly certain league does the same even disregarding the pro scene to such an extent that they will tell competitors theyâre not allowed to use a certain ability/hero or theyâd be penalized.
In terms of viewership I think the Esports worldcup will probably break a few records since it has the strongest participants whereas this recent HSC was mainly tier 2 Europeans and a few Koreans.
Itâs easy to have no convincing argument when youâre not willing to be convinced.
Again, youâre âsadâ for someone who earns 2x as much as you? Dude Iâm writing this stuff during my day in Asia. Youâre staying up all night arguing with someone on the internet when you should be going to bed literally 10 hours ago. Every single time I send a message yours follows within minutes man. Your Dopaminergic responses are completely out of whack.
Iâm not saying donât reply. Iâm not even telling you to agree with me. Let me have it all you want, but do during the hours of when the sun is up wherever it is that you live.
Well. No one is convinced because your argument is: avg mmr is lower with terran. Everyone in this whole sc2 community: yeah no sh!t, thats obvious. Bourne: âNAAAAHHHHHâ
I doubt that. But even if that was the case im happy for you. That someone with your abilities can earn that much fills me with joy, because then i know literally anyone can make it.
Look up my first reply today. It was at around 8 am in europe. On my way to work. At the office im allowed to do private stuff as long as i can complete all my duties at work, which i do. Some people attend english classes that you can sign in at work. I do pretty much the same here with you guys Look at eu time. Its 4:20 pm rn (blaze it).
So thank you that you are worried about my sleeping schedule or my income or career but i can assure you im doing fine. Whats next? You gonna make fun of my fitness? Of my height or my weight?
I wouldnt waste sleep on someone who is fuming about the fact that more casuals pick terran rather than zerg or protoss (which is obvious and logical for everyoneâŚwell except one guy it seems)
Do you think so? I donât. Hereâs why: they couldnât tell serral-level PvX already favored Protoss. They donât know how to collect & interpret data. Hereâs likely what will happen. Most of SC2âs players are from USA & western europe & korea. Itâs summer in those nations. Most nations have a summer break from school. That means SC2 popularity will be peaking late in the summer. According to Nephest, thatâs September-ish historically. EXCEPT, look at this chart: https://i.imgur.com/3uRQyN1.png. Weâve entered summer months & the game has already peaked. The peak happened 7 months early. So they are thinking their tournament numbers are doing âOkâ when in reality âOkâ numbers arenât good enough. The gameâs viewership should be moving up and up towards a yearly peak, but we already hit the peak.
I am doubtful. Remember that SC2 has to compete against other games for those viewers. Other games will be at the esports world cup. I think the proper way to analyze this event would be in what percentage of viewers, from the world cup, watched SC2, and to compare that to, for example, SC2âs normal percentage of esports viewership. I think the world cup will have higher numbers than normal, but relatively speaking SC2 is going to bomb. None the less, the balance counsel will feel like the higher than usual viewership is proof the game is saved, when in reality itâs a result of the Saudis throwing insane money at Starcraft and not the balance counselâs good patches. Based on the investment & expectations from that investment, SC2âs performance wonât be good enough.
TLDR, balance counsel will be confused about esports metrics in the same way they were confused about protoss performance metrics & this will give them the wrong read.
A bunch of B tier Protoss players: Nah
A bunch of pros, casters, commentators, and even balance coordinators: YeahâŚ
Again, Iâm sure you know more than they do.
I mean, my IQ is 144. Two posts ago you were crying about how mean I was and now youâre trying to insult me. *gasp I really am in awe that you would stoop so low.
I did. I noticed when you were messaging me. It was nearly 4 am in Europe.
You forgot occlusion array to determine if given wall can be rendered whole, partially or rejected completely. Every 3d engine stores level geometry in some sort of tree structure like BSP or octree and then draws each chunk of the level from near to far. This requires occlusion array to mark pixels which are already drawn on the screen to prevent drawing walls which are farther on those which are closer to viewing point. Iâm surprised you didnât mention it as it is a solid alternative to your depth test.
Batzy
A* is considered the most efficient and versatile pathfinding algorithm in any Euclidean space in most situations. Here you have good explanation:
h ttps://stackoverflow.com/questions/9511118/is-a-the-best-pathfinding-algorithm
So please try not to be smarter than you really are. Use standard and proven solutions rather than the crap your trying to âinventâ.
Itâs complicated. The most basic form is an octree where you cut the space in half in each dimension. Kd trees are sorta an extension of this but they allow the plane to move along the its normal axis. So they place the plane some distance along this axis to guarantee half the contents are on each side of the plane. BSP trees are similar to kd trees in that they split the world space by a plane but these planes are allowed to rotate to conform to the geometry in the space. So bsp trees are really common in enclosed spaced. Kd trees are better in open space. Octrees are better for lots of small objects packed into tight spaces because itâs kinda pointless to spend the computational power cutting the clusters in half over and over when they are all going to be on the screen anyway. My personal favorite is the octree mainly because it can be optimized to be extremely fast.
There is another one that is even faster and simpler than the octree that I use occassionally. Itâs called a spatial hash. You simply take the coordinate and divide it by some number, lets call it B, and round it to an integer. That integer is then placed into a map. Itâs very easy to compute if you are in the same region by simply doing hX = int(x/B) and plugging that into the map. The map is just an indexed array of lists. The index corresponds to hX and everything in the list at that index is in that space. The next question is how you decide if you intersect multiple spaces. Thatâs very easy with bounding boxes because itâs a simple greater than or less than operation for each axis. For circles itâs a circle to bounding box intersection test, which is also very easy to do. The reason spatial indices are useful is because they are more GPU compatible. So if you are doing GPU computing then spatial indices are a way to speed up the computation.
No, spatial partitioning algorithms are not an alternative to the rendering Z test. The rendering z test is simply a way for the renderer to decide what pixel is in front. It depends on what the renderer is trying to do because sometimes you want the pixel at the back. Spatial partitions are simply a way for the graphics engine, which is built on top of the renderer, to optimize the rendering. Itâs silly to send to stuff to the renderer if you know that stuff canât be seen, so why spend the GPU power on it. Thatâs why spatial partitions exist. The renderer runs on the graphics card and the graphics engine is built on top of that & runs on the CPU. They use spatial partitioning to reduce the amount of work the graphics card has to do & it also optimizes the usage of the PCIE bus but this is a nominal consideration on the new motherboards with PCIE 5.0+.
Nope, A* is an algorithm that does âOKâ at finding paths. The idea is to find the shortest path, but to know the shortest path requires a brute-force search, and those are far too slow for real-time. Even precomputing is too slow. It would take someone 3 days to save a map file if the paths were brute-forced and pre-computed. Finding a path is a matter of balancing two goals. The first goal is to reach the destination. The second goal is to move to closest available space that will take you towards the destination. So there is a short term goal and a long term goal. A* balances these two when considering options by weighting the distance from the current node to the new node, and the distance from the new node to the target destination, under the assumption that the path that has the lowest combined metric will be pretty close to optimal. The worst case scenario is that no path can be found, and that results in a brute-force search of all the options. To avoid that outcome, you have a maximum number of nodes that you are willing to consider before âgiving upâ. The way rts games handle that is it sends the unit to the node closest to the target that it did find.
A* is the industry standard but there are far better options. One of the obvious ones is to generalize the the walkable regions with a convex boundary. All the space within that boundary is considered walk-able. Then you find paths via ray tracing. Itâs way faster than A*. After you have the convex hull, you create a sphere that represents the largest sphere surrounding the target which doesnât go outside the walkable space. You shoot some random rays out from the unit. These rays are reflected off the normals of the convex boundary & you usually only have to bounce the rays around 3-5 times before they intersect the circle representing the target. You use the rayâs path as the unitâs path, except that you provide a post processing step so that the paths are more centered into the walkable space. That requires even heavier math, I am not going to bother explaining it. This is drastically faster than A* because the interior of the walkable space can be comprised of tens of thousands of nodes. The number of possible paths is in the trillions of trillions. But because you have the boundary you can find paths without caring about the individual nodes. You only care about the boundary. And donât think about copying this algorithm because I own 100% of the IP to it .
It really doesnt Help your Point that silver is 50% terran.
Really really really doubt that. Hey Look at me my IQ is 144. Im going to construct a forumula based on 1% Activity lead, AVG mmr and some other Magic and Boom every terran should have 25% more mmr, even Heromarine. Poof every pro terran is now Miles ahead compared to other races. Me so smoart.
Me so smoart im gonna use Casual Players to Claim my Race is Harder.
Me so smoart i say equal BMI means equal Body.
Me so smoart i say If you Split men and women into 2 height groups the systematic difference in height between men and women would dissapear.
Me so smoart despite everyone telling me more Casual Players tend to Pick terran im still confusing it with terran Players are more Casual.
Me so smoart im gonna Tell a Guy that i used to lift world record benches when i was 12.
Who do you wanna fool? You brag about everything. Your strength, income and intelligence. But If someone Looks a Bit into Detail suddenly its all gone.
You do you but i think you got some Ego issues.
Nope Not at all. For example right now its 7:40 am and my First Posts Yesterday were 23h ago and 1 day ago. A very reasonable time. No ? So how do you get 4 am Mr super smoart Guy? I think you dont get time zones either.
In this whole reply you did Not state any Argument. Cool.
Fine. But I can tell you that I started a business that allowed me to effectively retire in my early 30s. And it took a lot of analysis to do so.
Never said this.
You use the popularity argument when it boils down to all of 3 players.
Was an honest mistake. I did it at 14.
Now, THIS is a fair criticism lol. Even my wife would agree.
I didnât check it just now, dude. I noted it before⌠This isnât rocket surgery. And I didnât have to âunderstandâ time zones (but I was a pilot before, so, obviously know it) look up what time it was. I just asked siri what time it was in Berlin (itâs kind of the middle of Europe).
But i thought you are a pilot. Luckily not anymore it seems.
You said this. Where do you think i got this example from? you yourself made that point and it was funny as hell.
Sure And you âwerent even the strongest kid in your classâ
Wow. It would take every person like 2 seconds looking at some of your posts to figure that out.
So, yesterday you asked siri what time in berlin it is? And while it was around 8 o clock she said it was 4?
Dont you think you asked siri last weekend when hsc was going on? We obviously party through the night and i did post something in the nightimes when i got into my hotel.
So if you ask siri right now, how late is it in berlin?
I never was a professional pilot. I just flew recreationally. Stopped after Instrument rating. Iâll probably pick flying back up when Iâm in America, but when I moved to Asia I didnât want the hassle.
I was a year younger, so thgey wouldâve been 15. How much did I even say? I donât even recall. I benched 225 my Freshmen year of high school. I was 13 in the beginning of that year, while being 14 later in the year.
With all due respect but if your algorithm was so good and efficient, i GUARANTEE you that someone would have already stumbled upon it.
As far as my knowledge extends (iâm no mathematician, nor professional game dev, i create and develop CAD -like application very similar to AutoCAD) pathfinding algorithms depending on situation are:
A* - most standard
D*
D* Lite
Djikstra
You claim your algorithm can outperform these ? Really ? Of course i realize that these algoritms can be optimized so they can give slightly better results.
What ? Are you sure you donât confuse raytracing with raycasting ? These are two different techniques. Raytracing relies on projecting rays which bounce off objects multiple times representing light to generate photo realistic images and is usually computationally very expensive. In raycasting you project a single ray for each pixel only once to determine distance.
Nope. The skill required to do so is incredibly rare & having the time to work out all the kinks is even rarer. Some of these algorithms took over a decade to fully develop.
I have worked on CAD. Specifically it was a program that took AI generated designs and turns it into real objects via CAD algorithms. You say you are a programmer who works on CAD and then you turn around and deny the benefits of the algorithm I described and that tells me you donât know what you are talking about. You clearly donât understand the nature of the pathfinding problem & how to get around the obstacles that make the problem challenging to solve. I detailed a very unique & effective way to bypass the primary issue of the pathfinding problem which is the complexity of interior node configurations. Working on the boundary instead of the interior is faster because there are fewer boundary segments than interior nodes & because the complexity scales exponentially even a small decrease would be massive but itâs a decrease of factor of 10. But youâre not sold. I donât think you are a programmer. Nice try, though. I am detecting some Wikipedia copypasta. Think of it this way. The circumference of circle grows with the radius as 2PIr. The area inside the circle grows exponentially (PI*R^2). So youâre taking an exponential problem and turning it into a linear problem. Do you understand now.
I can talk about the high level concepts because the devil is in the details. Itâs kinda like saying âRocket goes up!â when describing how rocket engines work. You have chemical potential energy that creates a pressure differential and that produces thrust that makes the rocket move. Thatâs totally accurate but of course you couldnât build a working rocket with that information. The above explanation is drastically simplified & someone couldnât copy it without a 10 year time investment. Good luck to anyone who tries.