Athlonxp and Multiplayer

[CeltiberoMordred]

If anyone has an Athlonxp processor and tried to play multiplayer games, please report here your experience: if you were able to play, performance, lag, desync, ctd's, etc.

Thanks in advance, because I am very worried about this issue.

[donbatti]

yea, lag in 1vs1 but playable.. but still

[PaolinoPaperino]

Duron 2600+
just laggy
I could play 3vs3 too, with no speed/lag difference between a 1vs1 game.

[NihilisticCow]

I have an Athlon XP 2000+ and have lag.... Lots of it... In theory playable but I'd prefer not to miss the next ice age. I'm thinking of changing my motherboard/processor to a Pentium about the same class, very annoying. It's not a graphics related lag, as I can rotate/move around freely and I have absolutely no lag when playing even huge battles against the AI. So it must be something related to synchronisation online I guess... It would be really nice if a CA person was able to say whether this would be fixed or not, so that I know whether I need to upgrade or not... I don't criticise them for using technologies such as SSE2 as times move on and computers become obsolete, but I just want to know whether I need to spend more money or not.

One thing though could I try a game against another player with an Athlon XP or equivalent? I'm just wondering whether that would work... My xfire is nihilisticcow and online tag RTK4Caradoc.

[Kalle]

Hello all :)

Cow, I also have amd athlon xp 2000+ (1.67 ghz). Maybe we should try a game vs eachother, however I have not ventured online yet, gonna finish this short campain as Venice first. As you say on sp there is no problem.

Kalle

[t1master]

pala already said, to paraphrase 'there are issues with xp chips, and as thus, they're not supported.' my read of his statement is that they are not looking back. the online requires the newer processor tech.

[CeltiberoMordred]

So overall it seems it's playable, but with some lag (which is not number of players related). It would be interesting to try, as Nihilist and Kalle says, using Athlon XP only and see if the lag remains.

Also, we could save replays and watch them to check if they are identical or there is any kind of hidden desync.

And yes, Palamedes said it's not supported, but my read of that statement is that they didn't have time to fix it before the release. I think they can fix it in a patch.

An official response would be appreciated, because I must know if my only chance to play multiplayer is buying another machine. :help:

[Spartanian]

I postet twice about my prob,but no ateactions in other treads..again..

AMD 5000 dual core
ATI x1950xtx
2GIG RAM

I tried all settings..from all min to all max but still the same result.
Cant play a lagfree 2v2..3v3 is unplayable..so anyone with a similar system got the same prob?

In sp can play up to 4000 units without lag but over 4000 units lag!

[CeltiberoMordred]

I postet twice about my prob,but no ateactions in other treads..again..

AMD 5000 dual core
ATI x1950xtx
2GIG RAM

I tried all settings..from all min to all max but still the same result.
Cant play a lagfree 2v2..3v3 is unplayable..so anyone with a similar system got the same prob?

In sp can play up to 4000 units without lag but over 4000 units lag!

Maybe it's not your problem but other's. The speed of the game will be the speed of the worst machine involved.

If your mates are running the game with high graphic options when they can't afford it, or if someone is using an AthlonXp, then the game won't work properly.

Or maybe the multiplayer performance is poor itself and needs to be improved, who knows.

My advice for Mp games is to set to low-medium graphic details (although you can play single player with everything at high), no matters which computer you have.

Regards.

[sweetzero]

I have an Athlon 4200 dual core, 7900GT and 2 gigi of ram. Experienced no lag as yet in 4v4

[NihilisticCow]

I have an Athlon 4200 dual core, 7900GT and 2 gigi of ram. Experienced no lag as yet in 4v4

It's only an issue with Athlon XPs, not later ones that support SSE2.

[Kenchi_Sulla]

Small spinoff from this post but I was thinking about the following:

Multiplayer lags because every single units action has to be calculated. This means that if you have 6000 troops engaged in melee you have to send the calculations for that battle (every single man on the field needs it's own small calculation) over to all the other players. This isn't to bad if you play one vs one, but if you play 3 vs 3 the CPU needs to do all those calculations (nothing to do with graphics) and send that data to the other pc's to synchronise. This means if one pc is having trouble keeping up with things the game will "lag" - the stutter you see is a pc falling out of line.

This also means that it is important to have a good graphics card (give the CPU a breather) and on top of that you need a good CPU/ram to process the calculations.

I don't know how they did it in MTW/STW but perhaps the battle calculations were decided on each pc and only the basic data was transferred through the net. Perhaps because of the weaker graphics the CPU had an easier time keeping up with things. I'm not a expert on these matters.

Basicly this means that unless CA finds a way to lower calcs (e.a. streamline or simplify code) the lag will remain unless everyone gets superduper puters.

[Puzz3D]

I don't know how they did it in MTW/STW but perhaps the battle calculations were decided on each pc and only the basic data was transferred through the net. Perhaps because of the weaker graphics the CPU had an easier time keeping up with things. I'm not a expert on these matters.
RTW/M2TW works the same way as STW/MTW. Only the player's commands are sent to the host, and teh host relays those commands out to all the machines in the battle. The battle calculations are carried out independently on each machine in the battle which is why a machine can desync. Periodically, the host checks the gamestate of each machine in the battle, and if it doesn't match the host's gamestate, the host will wait for the machine to catch up. If the machine doesn't catch up within a certain amount of time, that machine is dropped from the game.


Basicly this means that unless CA finds a way to lower calcs (e.a. streamline or simplify code) the lag will remain unless everyone gets superduper puters.
They already simplified the battle engine in RTW most likely just for this reason. I can't imagine even more simplifications being made. Besides, they can't make major changes in the battle engine now. The players may be able to find settings that improves the performance such as turn off grass or use shader 1 and stuff like that. We used to turn off fauna and smoke in STW.

[tibilicus]

Personley I have had little lag in MP. If it's an issue with graphics or computer power thats not CA's fault. technology moves on. ~;)

[Nikodil]

If it's an issue with graphics or computer power thats not CA's fault.

No no no, NO! There's only one reason for lag, and that's naive network programming (see the description by Puzz3D above). Ask yourself, how come there are FPSs out there handling 100 of people or more with little or no lag? How come lagging on one machine doesn't neccesarily inflict lag on the other machines? Answer: because the devs realized the naive approach isn't robust and doesn't scale. So they went for a better approach, where the game can automatically resync whenever lag och desync occurs. There's no reason whatever that the same methods can't be applied to a game like TW. Except for marketing focus and budget constrains of course.

[Spartanian]

Personley I have had little lag in MP. If it's an issue with graphics or computer power thats not CA's fault. technology moves on. ~;)


....ati x1950xtx
..AMD64 am2 5000 dual core
...2 gig DDR2 800 RAM

lag in 2v2 battles(8-20 fps)..technology moves on..:dizzy2:

(i tried many diff settings..all min all max ..always same result..sp and mp)

[Crandaeolon]

Ask yourself, how come there are FPSs out there handling 100 of people or more with little or no lag? How come lagging on one machine doesn't neccesarily inflict lag on the other machines? Answer: because the devs realized the naive approach isn't robust and doesn't scale. So they went for a better approach, where the game can automatically resync whenever lag och desync occurs. There's no reason whatever that the same methods can't be applied to a game like TW. Except for marketing focus and budget constrains of course.

Not quite true. Using a FPS/MMO like serverside method is probably easier, not harder, to implement from a networking standpoint than what TW games are using. The problem is, such approach is not viable for TW because of the huge number of calculations involved in a battle.

Most performance issues in TW MP are caused by players who have too high graphical settings, or whose processors can't handle the workload. In M2TW there's the SSE2 incompatibility issue with older Athlons that's further complicating things. (I have an older Athlon 3000+, and for that reason haven't bought the game. Probably won't buy it even if it gets fixed, there are vastly superior MP games around.)

[Nikodil]

The problem is, such approach is not viable for TW because of the huge number of calculations involved in a battle.

I don't quite follow. The number of calculations is not neccesarily a factor. In fact, smart distribution of the calculations can even improve overall performance. Or do you mean the number of objects? I found this old article (http://www-ece.engr.ucf.edu/~jza/classes/4781/DIS/project.html) from 1998 that claims:
"The largest DIS exercise, part of DARPA's Warbreaker program, had 5,400 simulated entities interacting in a single DIS virtual world."
and moreover:
"Beyond DOD, many people see a likelihood that distributed simulation capabilities may be commercialized as entertainment. The scope of such an entertainment system is hard to predict but conceivably could be larger than the DOD goal of 100,000."

[Crandaeolon]

To use the data rates provided by the article, averaging 250 bytes/sec per entity and assuming that each individual soldier is a single entity, a computer simulating 1,000 soldiers would require 2 Mbit/sec of network bandwidth. For comparison, my upload bandwidth is 512 Kbits/sec. I could run 250 individual soldiers and that doesn't take into account networking overhead (more likely number is 200ish.) And that's only movement and positioning data under less than stressful conditions if I understood the article correctly. Much more data would be needed for an actual simulation.

Also using that kind of system, every individual arrow and slingshot and whatnot would require their own data. With current connection speeds we could control a couple dozen soldiers per player under that kind of system, with connections taxed to the limit.

In short: not gonna happen.

[ELITEofWARMANGINGERYBREADMEN88]

I think I got a Althon XP and No matter what I do, I can't even Log on... I don't know what to do now.. I did everything, shut off Firewall, I got 8 MB Cable,etc... and still can't get on

[Puzz3D]

Run dxdiag in the run box and see what it says for processor.

[ELITEofWARMANGINGERYBREADMEN88]

Intel(R) Pentium(R) 4 CPU 2.80 GHZ

that what I see when it comes up m8..

[Crandaeolon]

I'm considering purchasing the game, but am worried about the SSE2 issue. Is there any consensus yet about how playable the game is under non-SSE2 systems? Also, does CA plan to fix the issue in a patch?

My specs are: Athlon 3000+, 1 gig PC3200 RAM, Geforce 7600 gt 256. Most modern games run decently.

[CeltiberoMordred]

I'm considering purchasing the game, but am worried about the SSE2 issue. Is there any consensus yet about how playable the game is under non-SSE2 systems? Also, does CA plan to fix the issue in a patch?

My specs are: Athlon 3000+, 1 gig PC3200 RAM, Geforce 7600 gt 256. Most modern games run decently.

No news yet.

Single player: everything ok. Medium-high graphic settings with a good frame rate.

Multiplayer: I can play 1v1 multiplayer without problems.

2v2 and above: It's hard to me to play: "possible connection problems" "player X without response", etc., all the time. And if I can start playing, lag is horrible (all graphic settings at minimum).

My specs are: Athlon 2400+, 1 Gb DDR 333, GForce 7600GS 256.

[CeltiberoMordred]

deleted

[CeltiberoMordred]

deleted

[CeltiberoMordred]

deleted

[Orda Khan]

I'm considering purchasing the game, but am worried about the SSE2 issue. Is there any consensus yet about how playable the game is under non-SSE2 systems? Also, does CA plan to fix the issue in a patch?

My specs are: Athlon 3000+, 1 gig PC3200 RAM, Geforce 7600 gt 256. Most modern games run decently.
Hello Crandaeolon
Your system should run the game OK IMO. I can tell you that Intel/AthlonXP mixes are able to play 2v2 with no apparent problem. Any of the issues posted regarding MP are happening to all

.....Orda

[Kronos]

Hello Crandaeolon
Your system should run the game OK IMO. I can tell you that Intel/AthlonXP mixes are able to play 2v2 with no apparent problem. Any of the issues posted regarding MP are happening to all

.....Orda

Well there's alot of other factors, which Athlon XP just magnifies such as ping, pc specs etc. I was playing a 1v1 with someone using a top end Athlon XP pc with terrible lag. However when he switched to his laptop in the 2nd game with a fairly weak intel processor that the athlon is better than (except for mtw2) we played with no lag whatsoever. He was in australia and I also played a 1v1 with someone in the uk like me with a very similar athlon xp with much less lag so it was playable.

So my understanding of it is that SSE2 is required however any slight problem like a ping of 200 compared to 100, slightly less pc specs, slightly higher graphics settings make a huge difference to lag. 1v1's should be playable with low settings and someone who's geographically close to you but I hear they're fixing it in the patch iirc.

[Orda Khan]

Well there's alot of other factors, which Athlon XP just magnifies such as ping, pc specs etc. I was playing a 1v1 with someone using a top end Athlon XP pc with terrible lag. However when he switched to his laptop in the 2nd game with a fairly weak intel processor that the athlon is better than (except for mtw2) we played with no lag whatsoever. He was in australia and I also played a 1v1 with someone in the uk like me with a very similar athlon xp with much less lag so it was playable.

So my understanding of it is that SSE2 is required however any slight problem like a ping of 200 compared to 100, slightly less pc specs, slightly higher graphics settings make a huge difference to lag. 1v1's should be playable with low settings and someone who's geographically close to you but I hear they're fixing it in the patch iirc.
I know there are other factors to consider where lag is concerned. However, having experienced lag free battles with known XP and Intel processors in the same battle answers the initial question. As for lag issues, they are indiscriminate or at least, appear to be, so I see no reason to 'worry' about an AthlonXP

.....Orda

[Puzz3D]

As for lag issues, they are indiscriminate or at least, appear to be, so I see no reason to 'worry' about an AthlonXP
CA said AthlonXP was unsupported for multiplayer. They wouldn't say that if there wasn't a problem.

[NihilisticCow]

Well I've an Athlon XP and I get consistent lag issues with everyone I play at locations close to me and with whom I never had issues previously in RTW. So I think there is a problem, but I don't expect CA to fix this, as they've said the Athlon XP isn't supported. So I'm just going to upgrade.

[Orda Khan]

I can only report what I have experienced and that is no lag in 1v1 or 2v2 and fully aware that both processors were in game

........Orda

[UglyandHasty]

I have an Athlon 3.7 Ghz with a 64 chip, and can host 2vs2 without lag. I can play 3vs3 without lag if the others all have good comp. Only played one lagless 4vs4 on 4 try.

[Orda Khan]

They wouldn't say that if there wasn't a problem.
They say many things. I disregard most of them and I consider that statement to be a cop out clause

......Orda