#Sponsored Matchmaking Nodes

33 messages · Page 1 of 1 (latest)

flat remnant
#

Allow players and competitive leagues to personally fund "Sponsored Nodes" premium, high-performance servers that are integrated directly into the public matchmaking rotation. Instead of the developers bearing the full cost of global infrastructure, the community helps "crowdsource" a stronger, more localized network.

Funding: Users pay a monthly subscription to host a "Personal Premium Server."

Integration: These aren't hidden private lobbies. They are added to the Global Rotation Pool. When a match is found in that region, the game can prioritize these high-performance, user-funded nodes.

Resource Shift: By offloading the cost of these nodes to the users, Slapshot can redirect its current server budget toward upgrading the "Default" server clusters, ensuring a consistently strong connection for everyone, regardless of who is paying.

To incentivize players to pay for the community's infrastructure, they receive high-visibility "Vanity" rewards on their specific server:

Custom Loading Screens: When players connect to a sponsored node, they see the sponsor's custom art, team logo, or social media handles.

In-Game Branding: Sponsors can customize the rink board advertisements and the center-ice logo. This is perfect for streamers promoting their channel or leagues promoting their upcoming season.

Server Identity: Matches on these servers could be prefaced with a notification: "Now entering [User’s Name]’s Pro Server."

Shifting the load from default servers creates a surplus of "now money" for Oddshot Games to reinvest into core game development and server infrastructure.

#

Let’s turn the community’s passion into the game's backbone. By allowing players to sponsor the hardware, we get a decentralized, high-performance server network where the sponsors get the fame, the devs get the financial relief, and the players get the lowest ping of their lives.

naive nova
exotic venture
flat remnant
# naive nova they have stated they already have the best servers its just how the game is bui...

I have to disagree. The physics engine in this game is uniquely sensitive, and the inconsistency between matches is palpable. You can have the exact same six players in back-to-back games: one feels perfect, and the next is nearly unplayable.

Either Oddshot isn't being fully transparent about the server health, or there’s a misunderstanding of how their infrastructure works. Take this server I tracked, for example: ec2-18-205-248-81.compute-1.amazonaws.com.

This is a standard AWS US-East-1 instance. While AWS is great for scaling, these are virtualized servers, meaning the game is sharing physical hardware with non-gaming apps. If another company on that same hardware suddenly runs a massive data task, your "physics tick" gets delayed by 2–5 milliseconds. In a game like Call of Duty, you’d never notice; in Slapshot, that 5ms hitch makes the puck teleport and your stick feel like it's stuck in mud.

Furthermore, these servers use standard consumer routing optimized for bandwidth (like Netflix) rather than raw latency (Physics). They are running on a 60Hz tick rate, but without dedicated hardware, that 60Hz is never stable. Switching to specialized providers like i3D.net, Edgegap, or OVHcloud would provide the bare-metal consistency this game actually needs to be enjoyable.

#

So I invited you back to share a more sophisticated response that is helpful @naive nova

naive nova
#

lets say you are correct people 1st would not do this monthly thing for better servers and most people would complain about how one game feels good the next is garbage because they didnt get a premium server. they have always said they have the best servers which i cant fight them on because i dont know and tbh they probably dont but they are also a small team and im assuming they are doing everything they can so more people play the game

flat remnant
# naive nova lets say you are correct people 1st would not do this monthly thing for better s...

Most people probably wouldn't, you are absolutley correct. I guarantee you more (especially with a better final proposal) than you think would. The complaining about not getting a premium server is a non-issue, it's basically the current state of the game with getting a terrible server vs getting a less terrible of a server.

The extra money they receive from the people supporting the premium servers could go into even more premium servers. you could even add a $5/month pass to get priority to available premium servers (No doubt a load of hardcore slappers would want this).

Bottom line is that my initial idea is a rough skeleton of a thought. There's clearly more ways for Oddshot Games to generate more reveneue and implement better servers. I'd put $1000 down that new players are far more likely to be converted into consistent long-term players who spend money on this game if the servers were sufficient.

#

Heck, I suggested players with items named after them get a custom title only they have to represent that item. Turned down without a thought by Evereon. It's such a small thing but it would encourage people to generate ideas for new cosmetics for them (in-turn more people purchase pucks to get these cosmetics). The decision making here is honestly questionable.

exotic venture
#

@violet igloo is going to carpet bomb your home

flat remnant
#

So be it. I hope that in my passing everyone has a more enjoyable SS experience because of this information.

violet igloo
# flat remnant I have to disagree. The physics engine in this game is uniquely sensitive, and t...

Hi, thank you for your suggestions. Unfortunately what others have mentioned is correct and I can assure you we are being transparent. The subject is understandably a difficult concept and requires in-depth knowledge to fully grasp. Here are some facts I want to highlight:

  • We do run on bare metal 95% of the time. We only scale into the cloud when bare metal is at capacity.
  • We run at 120 tickrate
  • The servers are top of the line

The inconsistency you experience is because we don't have the lag compensation techniques other games do because the gameplay does not support it, so you feel the full inconsistency of the public internet.

This is not a money problem. If it were, it would have been resolved ages ago for sure.

#

I'm not sure where you're getting your information from (LLM?) but you highlight switching to specialized providers with dedicated hardware, but then mention 3 completely different companies with different usecases. We already use OVH as part of our bare metal stack, Edgegap is mostly focused on cloud, not bare metal (they are working on that though) and the only one relevant for your ask would be i3D. We use their direct competitor at the moment. i3D also scales into the cloud when bare metal is at capacity. They use hardware that performs worse than the hardware we currently use (we require a higher clock speed on the CPUs than they provide)

#

In conclusion, while I appreciate the passion for the game and your intent, the premise is unfortunately incorrect. I'm happy to talk about this in more depth but I suggest you do a deep dive into how lag compensation works, what use cases different providers have, and how latency works through the public internet before making unfounded accusations. I understand the frustrations, I really do.

flat remnant
#

I provided options that would all work better than a virtualized cloud network (the current server I was on 6PM EST). The accusations made were more directed at the comment I was replying to. That's my fault-I need to communicate batter (and be less aggressive).

However, there is still an obvious issue with the servers.

4:00AM EST I did get into the premium servers you described, awesome.

Looking at the server locations that are used for each region I'm starting to get a sense that maybe the issues is not the quality of the serves but the priority of servers during load times.

6PM - 7PM EST (02/25/2026) - there was a height of 130 active players playing.

4AM - 5AM EST (02/26/2026) - there was a height of just under 100 players playing.

Given that 6:00 PM EST on a Wednesday is a peak window for NAE and NAC, wouldn't it make more sense to prioritize bare-metal availability during those hours? It felt counterintuitive to get the highest-quality performance at 4:00 AM when traffic is lowest, while getting what felt like overflow routing during the most active part of the day.

Is this a known load-balancing issue you could look into? Specifically, would it be possible to implement a more intelligent server selector that prioritizes server quality and hardware specs (like the Game 2026 bare-metal nodes) during peak hours, rather than just optimizing for queue times?

Thank you for thoughtful reply.

#

Also, what is the context of the 95% threshold? Current active players or an average population ceiling?

leaden musk
#

mikolaj would be a harvard student if he wrote his essay with dedication like this

violet igloo
# flat remnant I provided options that would all work better than a virtualized cloud network (...

I'll use an analogy to explain how and why the "95%" works. Say there are two very large cities that are connected through a trainline, each line's capacity is 5 trains per hour, each train holds 100 people. The trainline is the best and easiest way to get between the two cities. Most of the day, there are way less than 500 people going from city to city so they can all comfortably fit on the trains. However, when there are a LOT of people during rush hour, the demand exceeds the capacity and busses have to be deployed to meet that demand. (the analogy isn't fully correct because the cloud servers are similar in performance, the analogy only extends to why we scale into cloud)

#

So the "overflow" system only kicks in when there are a LOT of people online, because the base system does not have the capacity to serve the amount of players. That being said, the overflow servers are on the same type of hardware. Virtualised or not, does not matter. The only difference is that the routing to those servers will be different because they're in a different (but close) physical location

#

A lot of games use Amazon servers for all of their servers (look into Gamelift for example)

#

Another thing to note, is we do not directly decide which data centre a match is deployed to. We ask our provider for a server through their API and they give us one. We do have a priority agreement with the provider to deploy on bare metal when there is capacity to do so, with a priority for certain locations to make latency as consistent as possible between matches. Only when it's physically impossible to do this does it scale up into the cloud. Overprovisioning bare metal to meet peak demand during all times of the day is something no provider would reasonably do if they're the ones bearing the provisioning.

#

If you link me to a match you were in, I can tell you whether it was bare metal or cloud. We only very rarely scale up into cloud servers.

#

Also note that i3D and Edgegap use the same provisioning logic as described above, it is not unique to our host

#

OVH is different because they don't provide any gameserver provisioning services, you just pay for the server and get a login so there is zero game orchestration happening with them directly. Providers like i3D, our current provider, and I believe in some cases Edgegap etc use OVH as part of their orchestration services with their own software

flat remnant
#

That makes sense so I hit the Hard Cap of the Bare Metal capacity during the Wednesday rush. Out of curiosity, what's the current player capacity for the NAC/NAE Bare Metal bucket before it starts spilling over into the cloud? It would be interesting to see how much more peak growth the premium servers can handle.

violet igloo
flat remnant
#

Also, could the fact that some players are using a 2000hz + polling rate and players having a mouse plugged in with a controller be causing stutter issues for everyone?

violet igloo
#

Unity can struggle with 1000+hz sometimes but it only affects your own client

flat remnant
#

that makes sense. Thanks for the responses dude, you were great.

tall ether