Page 2 of 2

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 09:41
by Rotonen
jaxad0127 wrote:
DarkWater wrote:no kill stats
Never.
Kill stats, especially "exp per time unit", are really rather important of a tool for power players, which are an MMO crowd which forms the hard core of the game world: the people who manage to get to the highest available players.

4144 is a usability pioneer for the client who actually hangs out on the field listening to the people who actually play the game. You should not discard everything at a first glance. Do try to evaluate things deeper than they seem at first. Convenience is powerful, it makes players stay longer and get frustrated far less.

The pickup of items is a rather fine example of design stupidity: if the server allows something, the client should do it. If you want to change it, fix the server.

The ability to follow a player is important too, IMO.

Looking into how the 4144 AFK implementation works should not hurt either.

Not that manasource devs would have the time for this, but I would recommend spending a full weekend with both clients. Do a session with 0.0.29.1, mana nightly, 4144. Then repeat 0.0.29.1 and mana nightly. A long session would for me mean about 8 hours of gameplay, but I also know that would be impossible to fit in. If anyone wants to go for this, I think I could have some time in early April to join in on such a test. :)

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 13:45
by Jaxad0127
Rotonen wrote:Kill stats, especially "exp per time unit", are really rather important of a tool for power players, which are an MMO crowd which forms the hard core of the game world: the people who manage to get to the highest available players.
EXP per hour, sure. "Kill X snakes to reach the next level", though....
Rotonen wrote:4144 is a usability pioneer for the client who actually hangs out on the field listening to the people who actually play the game. You should not discard everything at a first glance. Do try to evaluate things deeper than they seem at first. Convenience is powerful, it makes players stay longer and get frustrated far less.
I'm not discarding stuff at first glance. I'm discarding stuff that looks to be written and hard coded for platinum.
Rotonen wrote:The pickup of items is a rather fine example of design stupidity: if the server allows something, the client should do it. If you want to change it, fix the server.
I'm not against allowing you to pickup within a tile of where you are. I might fix eAthena to restrict the range again. 8 tiles is rather much, IMO.
Rotonen wrote:The ability to follow a player is important too, IMO.
And it works fine already.
Rotonen wrote:Looking into how the 4144 AFK implementation works should not hurt either.
Indeed.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 15:17
by thedarkfinder
Rotonen, jax has it out for dark water, always has. Every time she enters any thread, he comes right behind her, and threatens to close it, even if closing the thread goes against the rule of the forums.

He does not really take her serous.

If you want a good client, 4144 is a good client. But you want a good base, which is fine, fix the lag issues then hand it to 4144.


The one and only thing I can say. If we have a client issue with the standard client, we come here and post it. Jax screams it is either 1 a feature, or 2 already fixed.

If we have a issue with the 4144, we contact him in game, he comes to us, and does his damn best to recreate what is going on, and tries to fix it.

Which is where we are going to part.

4144 tries to help the people that use his clients.

Jax runs around threatening and dismissing people.

That is fine. No hard feelings on this side.

But dark water is dead on with her problems with the new client, the lag is bad.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 15:21
by thedarkfinder
Leela wrote:when i do what is explained in the first post, to test that, will i have a second client or will it change the one now? Because I like my client as it is now.

@DarkWater: Your critic is not based on the game but on a patch....
Her criticism was based on one version of the client compared to another version of a client. The asked us to play eA and see what we liked and disliked about it.

When you test the client. Install he second client into a different directory, like c:/program files/mana1 compared to c:/program files/mana. Then use a desktop shortcut to change between the two.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 15:27
by Rotonen
I've yet to experience any difference in lag between the clients. (Well I've yet to experience lag with the game. Probably my ISP and routing to Platinum are too good to run into that.)

It might be that something 4144 did makes the borderline situations slightly better, or then there is some sleigh of hand at play to smooth things out from the harsh reality. Either way, customer satisfaction seems to be greater with the 4144.

And to verify, by lag you do mean network based lag? Low FPS stuttering and skipping is a different problem. The difference is really essential for the related people here to actually fix things.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 16:27
by iceslice
Nice fighting. At least informative.
Dark Water has been brutal but in a +ve manner. :D

Well it's apparently true that ppls' trends to move from general client to 4144 client is higher than the opposite direction.
I would rather say, 4144 client is on it's way, general client on it's own way.
General client got to think of all kinds of server while 4144 is trying to include features that general client lacked.
Everyone is working for open source, no need to go after credits only. :D

One point, I have never seen 4144 to talk in public. Is this char backed by multiple persons ? :D (4144, dont mind)

EDIT: I have my vote for the plug-in/add-on feature.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 16:40
by thedarkfinder
Rotonen wrote:I've yet to experience any difference in lag between the clients. (Well I've yet to experience lag with the game. Probably my ISP and routing to Platinum are too good to run into that.)

It might be that something 4144 did makes the borderline situations slightly better, or then there is some sleigh of hand at play to smooth things out from the harsh reality. Either way, customer satisfaction seems to be greater with the 4144.

And to verify, by lag you do mean network based lag? Low FPS stuttering and skipping is a different problem. The difference is really essential for the related people here to actually fix things.
All of us, the family, run through a VPN, which runs from a fiber connection.

All of us report the same thing. If we are using the nightly we lag. The server just does not get us.

If we are using the 4144 we seem to lag less. But it is a subjective thing. Because the game server does lag.

I did a test in which I had the media server here at the house open two clients. One nightly, 4144, and I looked for lag. It was noticeable.


The media server is a quad core, with 3 way sli, and 24 gigbytes of memory.
It will run two cases of grand theif auto 4 on two screens, maxed out with out even getting warm.

And there is noticeable lag on the nightly compared to the 4144.

It is not the connection and it is not the computer. I can most of the time pull down 10mbps and push up 5mbps vpn server cap on connection. Most of the time.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 16:41
by thedarkfinder
iceslice wrote: One point, I have never seen 4144 to talk in public. Is this char backed by multiple persons ? :D (4144, dont mind)

EDIT: I have my vote for the plug-in/add-on feature.
He is really shy, but if you put a bag hat on and say "boo" alot. He comes around.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 16:47
by Rotonen
thedarkfinder wrote:I did a test in which I had the media server here at the house open two clients. One nightly, 4144, and I looked for lag. It was noticeable.
Any way to capture video of this difference with annotations or a walk-through description? Might help us track the cause. Wiresharking to see which sends/receives more or less packets could also hint for developers where to look for the crucial difference.

Does this manifest on 0.0.29.1 or has it been introduced later?

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 17:27
by DarkWater
Rotonen wrote:
thedarkfinder wrote:I did a test in which I had the media server here at the house open two clients. One nightly, 4144, and I looked for lag. It was noticeable.
Any way to capture video of this difference with annotations or a walk-through description? Might help us track the cause. Wiresharking to see which sends/receives more or less packets could also hint for developers where to look for the crucial difference.

Does this manifest on 0.0.29.1 or has it been introduced later?
well ask if I can.

The .0.29.1 does not seem to have the problem.

It does seem to be more of a problem with the social tab open. Lag on both clients.

But we are in portland oregon. So we could not be any more distance from the server.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 17:34
by Crush
I don't find it unlikely that 4144 will apply his changes to the 1.0 version as soon as it is released so that there will be a 4144 version of 1.0. But I would still be interested in the opinion of other players regarding which features should be ported to the mainline and why.

By the way, in my personal opinion the following features definitive no-gos:

Kill stats:
We shouldn't embrace the nature of TMW as a grinding treadmill.

Quickspell bar:
Too specific to eAthena. But I would be all for merging the smiley bar and the item bar in a general all-purpose quick bar which can also be used for any chat macro (including spell invocations) and has multiple pages which can be cycled.

Online list:
The method used by 4144, using http to download the online list dumped by the server at regular intervals, is just unfeasible. When we do something like that we should look into getting this info directly from the server and not via such a cumbersome hack.

But there are other features you can surely persuade me to support or even port myself.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 04 Mar 2010, 17:59
by DarkWater
Crush wrote:I don't find it unlikely that 4144 will apply his changes to the 1.0 version as soon as it is released so that there will be a 4144 version of 1.0. But I would still be interested in the opinion of other players regarding which features should be ported to the mainline and why.

By the way, in my personal opinion the following features definitive no-gos:

Kill stats:
We shouldn't embrace the nature of TMW as a grinding treadmill.
But it is, regardless of what you do it is. Kills stats over the hours shows the fallacy of the graveyard.

If you are not grinding for levels you are grinding for ore, fur, money.

Hey that is a great idea, increase kill stats to include drop stats. How often something is dropped over time.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 05 Mar 2010, 01:40
by DarkWater
Could the nightlies please at least have a ping counter. Please. It is easy to compare lag if you can know the ping rate to the server.

I have been doing some further testing. And would love to get some real data, other then just visual.

You can just copy and paste the 4144 code. That would work.

Re: Testing Manasource nightlies - 1.0 is approaching

Posted: 17 Mar 2010, 14:21
by Crush
New release candidate:
http://manasource.org/downloads

We incorporated a lot of new features in the last weeks. - some inspired by the 4114 client and a lot of own:

-Improvements at server select dialog
-Visible map portals in form of particle effects
-Option to show gender near player names
-Unassign button in setup keyboard tab.
-Update of translation files and improvements at the translation system
-Skill and Special buttons are not shown when the server doesn't make use of them
-Improved branding capabilities

Kudos to Jaxad who is the most active Mana client developer in the last weeks.