EPTJust to point out, to me boring mech is pretty much the swarm host era and more dynamic mech is what Nathanias does on stream where he is the aggressor and moves out quite a lot to poke and trade with cyclones hellbats and tanks. Seems to work for him in GM.
David Kim / Tim Morten Interview @ BlizzCon: Changes for S…
| Forum Index > SC2 General |
|
-IAEVAI-KolosS
Canada60 Posts
Just to point out, to me boring mech is pretty much the swarm host era and more dynamic mech is what Nathanias does on stream where he is the aggressor and moves out quite a lot to poke and trade with cyclones hellbats and tanks. Seems to work for him in GM. | ||
|
saalih416
19 Posts
| ||
|
ihatevideogames
570 Posts
| ||
|
KeksX
Germany3634 Posts
On November 15 2016 21:56 ihatevideogames wrote: So, let me get this right. A free to play game can have thousands of people watching in-client, both for tournaments and for high-ranked ladder matches, and a 40 euro huge AAA game from one of the most famous developers ever, that markets itself as the 'l33test' of esports, can only do what is basically a glorified in-client Twitch implementation? Technology is a bitch sometimes. You could also argue this about LoL with replays. Here's the problem with the watch feature: Other games can simply send gamestates to your client and your client does not care what happened in the past, just what is happening right now. StarCraft II, because of its architecture, needs to re-do EVERY step in order to get a valid state out of it (this is why you can't jump in replays, just fast-forward really fast), and providing a watch-system that offers a nice user experience will be a bitch to implement. And I'd say that right now there are bigger issues than this. | ||
|
Tuczniak
1561 Posts
David Kim: And as for skins, I think we’ve said before that we wanted to do it, but performance issues made it so that we can’t do it, but our engineering team has solved those issues. The technology is here!Watch tab in-game Video is fine. But ingame observing buffered through server with delay is possible on SC2 engine. Although I don't consider it main priority for SC2. Twitch experience is pretty good already.So say we’re 95% there after these changes go in, then obviously there won’t be a major patch next year, but if we’re halfway there, then maybe we do. It really depends. It's scary that they would even consider not doing big patch next year. These changes are nice, don't get me wrong. But even if all of them make it through the game would 100% greatly benefit from big unit changes next year. Ideal would be at least one patch like this half way through 2017. | ||
|
eviltomahawk
United States11135 Posts
On November 15 2016 22:18 KeksX wrote: Technology is a bitch sometimes. You could also argue this about LoL with replays. Here's the problem with the watch feature: Other games can simply send gamestates to your client and your client does not care what happened in the past, just what is happening right now. StarCraft II, because of its architecture, needs to re-do EVERY step in order to get a valid state out of it (this is why you can't jump in replays, just fast-forward really fast), and providing a watch-system that offers a nice user experience will be a bitch to implement. And I'd say that right now there are bigger issues than this. This is also why Heroes of the Storm has such a mediocre reconnect feature. It essentially tries to resume from replay then catches up to the live game state at the very end of that. It ends up being a slow, cumbersome process. Fortunately, games don't go for too long in HotS so the reconnect times aren't even longer, but unfortunately those lost minutes are even more important in those short game times. I think that the issues in both SC2 and HotS with these features show how hard-coded these problems are. The engine may not have been built from the ground-up to accommodate for those features when they started working on it nearly ten years ago, and considering how convoluted Blizzard's legacy code can become in their other games, I wouldn't be surprised to see similar coding difficulties in the SC2/HotS engine. | ||
|
mantequilla
Turkey781 Posts
I kinda swore I won't buy lotv for that price (40 eur) unless they make the gameplay noticeably fun, and not frustrating. Better to subscribe to a streamer with that money for 9 months since watching is still some fun. If they suck it up and make a huge discount maybe another wave of players will buy it, but blizz never discount do they? | ||
|
Hider
Denmark9433 Posts
So say we’re 95% there after these changes go in, then obviously there won’t be a major patch next year, but if we’re halfway there, then maybe we do. It really depends. Lul 95% there? This patch at best increases strategic diversity potential from 10% to 20%. Game could be a million times better. | ||
|
xAdra
Singapore1858 Posts
On November 15 2016 21:56 ihatevideogames wrote: So, let me get this right. A free to play game can have thousands of people watching in-client, both for tournaments and for high-ranked ladder matches, and a 40 euro huge AAA game from one of the most famous developers ever, that markets itself as the 'l33test' of esports, can only do what is basically a glorified in-client Twitch implementation? This is incredibly puzzling to me as well. DotA has great spectating tools already and it's free to play. This game costs so much money and is so big budget yet is only beginning to think about implementing this. | ||
|
paralleluniverse
4065 Posts
On November 15 2016 05:49 TeamLiquid ESPORTS wrote:One thing teased in the announcements was the idea of a new Watch tab in-game. Can you talk a bit more about how that will function? Tim Morten: We’re working towards a future where video content can be consumed directly within the Starcraft 2 client. We’re not at a point yet where we can predict when that’s going to land next year, but we are working towards being able to host esports video content within the client. So is that going to be a direct video feed, or would you possibly be able to observe within the game itself? Tim Morten: What we’re envisioning for now is a video feed. I think that observing on a broad scale, where we are connecting potentially hundreds of thousands of players to a game has a lot of implications in terms of performance, bandwidth, the potential for those observers to disrupt the game, which we absolutely do not want to happen. So video is what we’re focussed on for now. Hype went up. Then went down. Video is a tiny first step, it should really be in-game streaming, like Dota 2, not video. | ||
|
paralleluniverse
4065 Posts
On November 15 2016 22:48 mantequilla wrote: well good little features they are adding I kinda swore I won't buy lotv for that price (40 eur) unless they make the gameplay noticeably fun, and not frustrating. Better to subscribe to a streamer with that money for 9 months since watching is still some fun. If they suck it up and make a huge discount maybe another wave of players will buy it, but blizz never discount do they? http://us.battle.net/sc2/en/blog/20372288/starcraft-ii-battlechest-and-legacy-of-the-void-price-drop-11-10-2016 | ||
|
gab12
Poland147 Posts
 ) | ||
|
lestye
United States4208 Posts
On November 15 2016 23:43 paralleluniverse wrote: Hype went up. Then went down. Video is a tiny first step, it should really be in-game streaming, like Dota 2, not video. That'd be nice, but I don't think that's possible with the SC2 client, since its so old and its not really designed around it. | ||
|
KeksX
Germany3634 Posts
On November 15 2016 23:18 xAdra wrote: This is incredibly puzzling to me as well. DotA has great spectating tools already and it's free to play. This game costs so much money and is so big budget yet is only beginning to think about implementing this. You're approaching this from the wrong angle. Just because one game is free to play doesn't mean it didn't cost any money to develop. Also, even though the game have similar controls, from a networking perspective they're completely different. DotA2 uses dedicated servers, StarCraft II uses p2p servers. Meaning for DotA2, there is a server that handles all the server related stuff(updating time/physics, giving commands to NPC units), and players then only push input to those servers and get updates from the server on what the other players did. 10~15 units or so, not a lot of data. Compared to StarCraft II where this is done by a player and also with a lot more data and with a verification method to ensure that every player at all times has the same state (and that the host can't cheat the state). I'm missing a lot of technical details to make the explanation easier. The point is, networked RTS are really hard to do and take a lot of engineering power. SC2 was designed without all of these fancy features in mind, DotA2 has fewer problems from the beginning and also was designed with all these problems in mind. This doesn't mean that DotA2 had it easy, it had it's own share of problems since the Source Engine wasn't built with RTS in mind. But talking about all those fancy customization and network features - those are definitely harder to implement in StarCraft II. This is no excuse of course to not try anyway for Blizzard, since I think if there's a company that can push the technical boundaries for this, it's them. But as I said earlier, it was probably not a high priority feature so far. On November 15 2016 22:42 eviltomahawk wrote: This is also why Heroes of the Storm has such a mediocre reconnect feature. It essentially tries to resume from replay then catches up to the live game state at the very end of that. It ends up being a slow, cumbersome process. Fortunately, games don't go for too long in HotS so the reconnect times aren't even longer, but unfortunately those lost minutes are even more important in those short game times. I think that the issues in both SC2 and HotS with these features show how hard-coded these problems are. The engine may not have been built from the ground-up to accommodate for those features when they started working on it nearly ten years ago, and considering how convoluted Blizzard's legacy code can become in their other games, I wouldn't be surprised to see similar coding difficulties in the SC2/HotS engine. Good point! This is why I was hoping for Heroes to be a financial success, since every feature they'd implement there could be ported to StarCraft II with relative ease. | ||
|
Topin
Peru10116 Posts
| ||
|
Incognoto
France10239 Posts
| ||
|
Cele
Germany4016 Posts
| ||
|
Edowyth
United States183 Posts
We almost never see mech play in LotV—that’s something we obviously wanted to address—and SkyToss is another thing that’s going to be changing—we’re changing three stargate units. That’s our ultimate goal, and hopefully this major patch pushes us towards that. ... I have no idea what he's talking about here. Does anybody see a viable composition for SkyToss coming about due to these changes? It seems like, to me, even with mass storm + cannons, SkyToss only became less likely in this patch. Here's what I think: - Carriers losing release interceptors decreases the slim chances they'll ever be used - Carriers having 5-mineral interceptors doesn't affect the rapidity with which interceptors die (so, you can afford more of them, but they still do nothing if the opponent plays well) - Tempests costing 50% more supply makes them even less likely to be used - The ability doesn't nearly compensate for the loss of range for anti-ground (again making the unit less useful than previously. Anti-ground range could kill marines, this ability won't hit marines.) - Void speed is a token change. The unit still plays exactly the same, they're just very slightly faster now. Maybe it's just PR speak? Even counting extreme-late-game mass-cannon + storm and vs-mech, it seems like air units will be generally less useful than previously. | ||
|
Clbull
United Kingdom1439 Posts
Ever since Heart of the Swarm, I've personally disliked the direction Blizzard took with the game by buffing harassment options for all three races. At one point, certain harassment options were so overpowered that if you didn't build specifically to counter that option, you'd lose the game outright. In this expansion however, not only did they cut minerals per base to make players mine out quicker and force aggressive expansions, but they also buffed harassment options even more. Adepts, Overlords, Ravagers Siegeivacs and Warp Prisms are some of these buffs that have just made the game so bad. The Siege Tank buff is a welcome change, but that's about it, really. The rest of the changes have been dubious at best. Besides, if the complete lack of playerbase on the test ladder is any indication, 2017 will be the final nail in the coffin for SC2. | ||
|
Clbull
United Kingdom1439 Posts
On November 16 2016 01:49 Edowyth wrote: > We almost never see mech play in LotV—that’s something we obviously wanted to address—and SkyToss is another thing that’s going to be changing—we’re changing three stargate units. That’s our ultimate goal, and hopefully this major patch pushes us towards that. ... I have no idea what he's talking about here. Does anybody see a viable composition for SkyToss coming about due to these changes? It seems like, to me, even with mass storm + cannons, SkyToss only became less likely in this patch. Here's what I think: - Carriers losing release interceptors decreases the slim chances they'll ever be used - Carriers having 5-mineral interceptors doesn't affect the rapidity with which interceptors die (so, you can afford more of them, but they still do nothing if the opponent plays well) - Tempests costing 50% more supply makes them even less likely to be used - The ability doesn't nearly compensate for the loss of range for anti-ground (again making the unit less useful than previously. Anti-ground range could kill marines, this ability won't hit marines.) - Void speed is a token change. The unit still plays exactly the same, they're just very slightly faster now. Maybe it's just PR speak? Even counting extreme-late-game mass-cannon + storm and vs-mech, it seems like air units will be generally less useful than previously. Blizzard have taken entirely the wrong approach with the Tempest. Yes, Tempests were too supply efficient, but their range wasn't the issue, and were actually a core design principle of the unit when they were first introduced in HotS as an Infestor/Brood Lord counter. If Tempests didn't exist, Protoss couldn't have survived those 3 hour stalemates you always saw from Swarm Host PvZ games. The fact that Tempests hard-counter Mech says more about how crap Mech anti-air is. For anti-air, Mech only has two options, Thors which are terrible against capital ships and don't even perform that well against swarms of air units thanks to a trick known as magic boxing, or Cyclones, which were a bad idea in both iterations. There is a reason why Terran relied on mass Vikings as Mech anti-air, and that is because the other options outright sucked. A 20% splash radius increase won't be anywhere near enough to fix the Thor's flaws. | ||
| ||
