Client-side demos will never work.
- Thread starter T2A`
- Start date
-
Two Factor Authentication is now available on BeyondUnreal Forums. To configure it, visit your Profile and look for the "Two Step Verification" option on the left side. We can send codes via email (may be slower) or you can set up any TOTP Authenticator app on your phone (Authy, Google Authenticator, etc) to deliver codes. It is highly recommended that you configure this to keep your account safe.
Clientside demos are mandatory for participation in every ladder and league I know of, in every single FPS game used for competition online. They're often the crucial piece of evidence in catching and punishing cheaters.
According to anti-cheat staff, serversides don't usually work for this purpose due to problems with sync (never mind the logistical problems of accessibility and the performance hit on the server). In UT2k4, for example, some cheats discoverable through clientside demos could not be verified through serversides.
This is not trivial, since it appears that Epic has done little (if anything) for anti-cheat in UT3 and the game was already cracked open shortly after release. Without reliable clientsides, anti-cheat in UT3 is challenging...to say the least.
But how would you obtain it? cheaters like to infest unadmined servers afterall, so lest the server is recording non-stop (unlikely, waste of bandwith), your best bet is a bunch of clientside demo-rec's showing off the suspicious playstyle.
That is how it usually works, and allways has really, its circumstatial evidence, yes, but emass enough of it and it starts to paint the picture.
As for "Supported", its just PR Newspeak, its so delightfully vague a term that it could mean practically anything, whilst not carrying any direct negative connotation, so you can toss it brashly about and never have to worry about it coming back to haunt you later..
I would think that the number of people actively cheating on ladders is pretty slim in a community like UTs (at least once stuff like UTAN comes out). But I do get what you mean. Still, you have to get the client side demo of the person cheating to see that they are cheating. I just meant a server side would be better than me providing a client side demo of someone else cheating. Cheat-like activity can be attributed to anything from lag to connectivity issues in a client side demo not featuring the cheater.
Last edited by a moderator:
You could also spectate him and record a demo of him. So the demo is clientside on your side. Then you submit this demo to UTAN or alike for a global ban.
Works great. We at jolt had a sysetm that you can rename yourself to admin_needed and that gets reported to the IRC admin channel for these servers. Then some admin joins and records a demo.
Doesn't work great. Spectators have always been given less bandwidth allowance than real players, so what you're seeing isn't anything close to what a true client-side demo would show.
But you're still relying on network conditions between you and the server and him and the server.I made a short frag movie once.. On average I made a frag worth recording every 2 hours. So basically I recorded with demorec for a few hours before creating a new demo. In 2k3 that would get you a file of about 30 MB. I can tell you that if you d record 2 hours with fraps it will kill your fps in game, and on top of that, you ll get a file so huge no modern computer could survive recording a few of those (unless you want a crap quality vid).
Not being able to record clientside demos = no more frag movies. Thats just a fact. Yeah serverside still works, but only 1 in 100 players owns a server.
Last edited:
Where have you been for the last 3 years? Cheating is pretty common, not on the level of CS, but still there. UTAN does not stop cheating, all it did for 2k4 at least was allow for shared BAN lists and global bans, all easily bypassed. At TWL, all client side demos had to be recorded, and if an admin found reason to check, you had to turn in your client side demo, if you failed to, you got suspended. Client side demos are most accurate depiction of what the player is seeing.
I know what the rules are. That isn't what I was talking about at all.
UTAN would help in ladders because it would globally ban anyone caught cheating, even on public servers.
I know cheating on public servers is fairly common, but we aren't talking about public servers.
The spectate method works fine in pub servers, where most cheaters are either stupid or don't care to cover their tracks too carefully. In competition, tho, cheating was refined to a much subtler...um, art?
But yes, your point is important - clientsides are vital for anticheat not just in competitive formats, but also in pubs. There is in fact no other reasonable method for documenting pub cheating; enabling serversides 24/7 on a pub server is insanely impractical.
The problem is also settings for server side recording. Most of the demo tic rates are set pretty low to try to help performance, so they are choppy, and it gets harder to identify the noticeable tearing and chop of lockon. Most people don't use lockon in a competitive setting anyway, and you need to study other behavioral patterns to identify a cheat, which actually can be observed in most demos. They, though, are much less conclusive.
Honestly though, lockon is the only bot feature that is silly obvious, and only a couple option features made to hide it make high quality demos necessary. The rest of it requires behavioral study.
Honestly though, lockon is the only bot feature that is silly obvious, and only a couple option features made to hide it make high quality demos necessary. The rest of it requires behavioral study.
and like I said, where have you been for the last three years? Should I start to name off botters caught in sanctioned matches?
UTAN works when people are caught, but it does nothing to actually catch them, at least last time I checked, it is not an anti-cheat. If you can't catch them, how can u expect to ban them? UTAN does not globally ban unless there is good proof.